Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Wiki Index

Content catalog. Every wiki page listed under its type with a one-line summary. Read this first to find relevant pages for any query. Last updated: 2026-09-02 | Total pages: 158 (entities + concepts + meetings; 34 raw sources incl. 17 meeting notes)

Entities

Entities / organizations

  • frames - FRAMES, the IBC2026 Accelerator umbrella project hosting the film
  • ibc - IBC, the organization running the Accelerator programme (hosts FRAMES)
  • rai - RAI, Italian public broadcaster; archive source; FRAMES Champion
  • ebu - EBU, public media alliance; FRAMES Champion co-proposer
  • movielabs - MovieLabs, OMC author; FRAMES Champion co-proposer
  • streamline-media - FRAMES Participant; John Hobbs’ streamline-media.net company
  • amd - AMD, FRAMES Participant (with HP); AI compute hardware kit
  • hp - HP, FRAMES Participant (with AMD); Z-series workstations, kit delivery
  • eddie-ai - agentic AI editor; FRAMES Participant (aka ADI AI)
  • holli - HOLLI, media sustainability/Data-AI consultancy; FRAMES supporter
  • insa - INSA Hauts-de-France (Valenciennes), FRAMES Champion
  • mediadata - MedIAdata, the INSA x HOLLI partnership that develops hollm and participates in frames
  • aimici - responsible AI standards partner; designed the transparency framework
  • xrbb - Extended Reality Berlin-Brandenburg; FRAMES team member
  • etc-usc - Entertainment Technology Center at USC; FRAMES team member
  • itv - British broadcaster; FRAMES team member
  • enigmatic - FRAMES team member (distinct from eddie-ai)
  • gc-sc - The Global Creative & Security Community; FRAMES team member
  • press-play-labs - Shamir Alibhai’s org; FRAMES team member
  • bbc - British public broadcaster; FRAMES team member
  • yamdu - production-management / scheduling / call-sheet cloud tool (used by florian-reimann)

Entities / people

Entities / production

  • one-small-crawl - the film: IBC2026-SNAIL-001, animated short, 5 eras, hybrid archive + AI pipeline

Entities / production/deliverables

  • deliverables - the film’s output deliverables (final, YouTube export, score)

Entities / production/archives

  • artemis-ii - space thread: NASA mission + RAI news + “Eagle has landed”

Entities / production/characters

  • snail - Stella, CHAR-001, the snail protagonist walking Route 66

Entities / production/props

(empty - Nutella jar product placement is a candidate prop entity)

Entities / production/environments

  • route-66 - the road: Chicago to Santa Monica, per-scene geography

Entities / production/sequences

  • sc01 - 01 The Route 66 Sign (5 shots, seq sc01)
  • sc02 - 02 The Shoulder & the Diner (7 shots, seq sc02)
  • sc03 - 03 The Rain & the TV (10 shots, seq sc03)
  • sc04 - 04 The Decay & the Dream (22 shots, seq sc04)
  • sc05 - 05 The End (6 shots, seq sc05)

Entities / production/shots

Each shot page carries a “MovieLabs export (JSON)” section merged from the OMC export: generation brief v0.3.0 (all 50), shot list v0.2.0 revised / AGT-020 fallback, and per-shot OMC sidecar where present (34/50).

sc01

sc02

sc03

sc04

sc05

Entities / tools

  • hollm - AI production assistant (HOLLI/mediadata); manages the Production Brain
  • unsloth - open-source LLM inference tool (runs Ornith/LFM/Qwen for HOLLM)
  • ornith - LLM run via Unsloth for HOLLM
  • lfm - LLM run via Unsloth for HOLLM
  • qwen - Alibaba LLM run via Unsloth for HOLLM
  • kitsu - open-source production tracker (CGWire); shot list + review backbone
  • resolve - DaVinci Resolve edit/colour/delivery pipeline
  • premiere - Adobe Premiere Pro NLE
  • after-effects - Adobe After Effects compositing/motion
  • blender - open-source 3D creation suite
  • maya - Autodesk Maya 3D animation
  • unreal - Unreal Engine real-time renderer
  • topaz - Topaz Gigapixel/Video AI upscaling + enhancement
  • lora - LoRA (Low-Rank Adaptation) model fine-tuning
  • fal - Fal.ai generative media API
  • gemini - Google Gemini (LLM/vision)
  • chatgpt - ChatGPT (OpenAI) assistant
  • eddie-ai - agentic AI editor (aka ADI AI)
  • edit-mind - Edit Mind editing assistant
  • moments-lab - Moments Lab media intelligence
  • google-cloud - Google Cloud platform
  • graphdb - GraphDB knowledge-graph store
  • slack - Slack team communication
  • suno - Suno AI music generation
  • rai-crits - RAI CRITS tooling
  • render-youtube-final-py - Resolve render script for the final YouTube export

Meetings

Concepts

Raw sources

Comparisons

(none yet)

Queries

(none yet)

External reference

  • MovieLabs OMC export (raw data): D:\PRODUCTION\ONE_SMALL_CRAWL\Knowledge\movielabs-osc-export\

Wiki Schema

Domain

Knowledge base for One Small Crawl (production id IBC2026-SNAIL-001), an animated short built for the IBC Accelerator 2026 showcase. Hybrid pipeline: authentic archival footage retrieved via the ARC pipeline, AI-generated character animation and VFX, era-style enhancement, and sound design. Authoritative machine-readable source: the MovieLabs OMC (Ontology for Media Creation) JSON export received 2026-09-02 (extracted to ../movielabs-osc-export/).

Conventions

  • File names: lowercase, hyphens, no spaces.
  • Every wiki page starts with YAML frontmatter (see below).
  • Use [[wikilinks]] to link between pages (minimum 2 outbound links per page).
  • When updating a page, always bump the updated date.
  • Every new page must be added to index.md under the correct section.
  • Every action must be appended to log.md.
  • Strict ASCII policy: all wiki files are byte-clean ASCII. No em/en dashes, no smart quotes, no CJK/fullwidth punctuation. Fancy punctuation is replaced with ASCII (--, ->, ...). This is enforced because model output on this machine intermittently emits non-ASCII; the perl purge command is in the log.
  • Wiki pages use wiki cross-links in [[bracket]] form; raw-layer pages are type: source with absolute path frontmatter.
  • Provenance markers: on pages that synthesize 3+ sources, append ^[raw/.../source.md] at paragraph ends where a specific source supports the claim.

Frontmatter

---
title: Page Title
created: YYYY-MM-DD
updated: YYYY-MM-DD
type: entity | concept | source | comparison | query | summary
tags: [from taxonomy below]
sources: [raw/.../source-name.md]
confidence: high | medium | low
---

confidence and contested are optional but recommended. Lint surfaces contested: true and confidence: low pages for review.

raw/ Frontmatter

Raw sources use type: source, path: (absolute on-disk location), ingested:, and sha256: of the body (everything after the closing ---).

Tag Taxonomy

  • Production: production, project, deliverable, timeline, ibc2026, meeting, marketing
  • Media/layers: archive, ai-generated, ai-enhanced, composite, score, animatic, audio
  • Eras: silent, newsreel, kodachrome, vhs, digital, era-look
  • Ontology/tech: omc, movie-labs, ontology, schema, naming, pipeline, tool, unreal, metadata, compliance, transparency, kitsu, google-cloud, ai-editor, hardware, color, ves, inference, model, unsloth, ornith, lfm, qwen
  • Story: character, snail, route-66, narrative, artemis-ii, space
  • Team: team, person, rai, ibc, editor, agent, organization, ethics, frames, aimici, xrbb, etc-usc, itv, enigmatic, gc-sc, press-play-labs, bbc, holli, insa, open-source
  • Ethics: ethics, responsible-ai, transparency, disclosure, eu-ai-act, accountability, copyright, compliance, provenance, c2pa
  • Meta: experiment, demo, confidence Rule: every tag on a page must appear in this taxonomy. Add new tags here BEFORE using them.

Page Thresholds

  • Create a page when an entity/concept appears in 2+ sources OR is central to one source.
  • Add to existing page when a source mentions something already covered.
  • DON’T create a page for passing mentions.
  • Split a page when it exceeds ~200 lines.
  • Archive a page when fully superseded (move to _archive/, remove from index).

Entity Pages

One page per notable entity: overview, key facts and dates, relationships ([[wikilinks]]), source references.

Concept Pages

One page per concept: definition, current state of knowledge, open questions, related concepts ([[wikilinks]]).

Comparison Pages

Side-by-side analyses: what/why, dimensions (table), verdict, sources.

Update Policy

  1. Check dates – newer sources generally supersede older ones.
  2. If genuinely contradictory, note both positions with dates and sources.
  3. Mark the contradiction in frontmatter: contradictions: [page-name].
  4. Flag for user review in the lint report.

Source of Truth Hierarchy (One Small Crawl)

  1. Agent Production Bible v1 (raw/movielabs/agent-production-bible.md) – governance, agents, pipeline, hard constraints.
  2. Era & Visual Look Guide v001 (raw/movielabs/era-look-guide.md) – per-scene look; wins over Bible on look parameters.
  3. OSC Asset Naming Schema v1.0 (raw/movielabs/asset-naming-schema.md) – file naming.
  4. PeriodO crosswalk v001 (raw/movielabs/show-ont-periodo-crosswalk.md) – period gazetteer, research corpus.
  5. Raw source notes in raw/archive/, raw/rai/, etc. – on-disk assets.

Organizations

  • AIMICI: responsible AI standards partner; designed the transparency framework
  • AMD: AMD, FRAMES Participant (with HP); AI compute hardware kit
  • BBC: British public broadcaster; FRAMES team member
  • EBU: EBU, public media alliance; FRAMES Champion co-proposer
  • Eddie AI: agentic AI editor (aka ADI AI)
  • Enigmatic: FRAMES team member (distinct from eddie-ai)
  • ETC/USC: Entertainment Technology Center at USC; FRAMES team member
  • FRAMES: FRAMES, the IBC2026 Accelerator umbrella project hosting the film
  • GC-SC: The Global Creative & Security Community; FRAMES team member
  • HOLLI: HOLLI, media sustainability/Data-AI consultancy; FRAMES supporter
  • HP: HP, FRAMES Participant (with AMD); Z-series workstations, kit delivery
  • IBC: IBC, the organization running the Accelerator programme (hosts FRAMES)
  • INSA: INSA Hauts-de-France (Valenciennes), FRAMES Champion
  • ITV: British broadcaster; FRAMES team member
  • MedIAdata: MedIAdata, the INSA x HOLLI partnership that develops hollm and participates in frames
  • MovieLabs: MovieLabs, OMC author; FRAMES Champion co-proposer
  • Press Play Labs: Shamir Alibhai’s org; FRAMES team member
  • RAI: RAI, Italian public broadcaster; archive source; FRAMES Champion
  • Streamline Media: FRAMES Participant; John Hobbs’ streamline-media.net company
  • XRBB: Extended Reality Berlin-Brandenburg; FRAMES team member
  • Yamdu: production-management / scheduling / call-sheet cloud tool (used by florian-reimann)

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

AIMICI

AIMICI is a partner to the FRAMES project and has been working on standards around responsible AI best practice in industry over the last two years. It designed the aimici-transparency-framework and proposed the responsible-ai-stamp.

Identity

  • FRAMES project team member (Project Team - First Authors: Kathryn Webb and Ahsan Mallick, AIMICI).
  • Focus: responsible AI best practice, transparency, and standards for the media industry.

Role in FRAMES

  • Designed and tested the aimici-transparency-framework on the project (WHAT / WHO / HOW / WHEN / WHERE + three levels of records).
  • Proposed the responsible-ai-stamp (in development, Q4 2026 consultation).
  • Identified two key levers to help industry navigate AI concerns at scale: training and transparency.

People

AIMICI - responsible AI standards (no Wikipedia article; official site)

AMD

AMD (Advanced Micro Devices) is a US semiconductor company. With HP it is listed as the FRAMES Participant “AMD/HP” (the IBC2026 press release writes “AMD-HP”) supplying hardware for the frames proof-of-concept one-small-crawl.

Role in FRAMES and One Small Crawl

  • Supplies the compute side of the production’s hardware kit; the crew’s consolidated order is the “AMD HP” hardware order coordinated by roberto-iacoviello under a June deadline and tracked with HP’s Rick.
  • The AMD Strix Halo processor powers the HP Z2 mini workstation that runs hollm for the IBC Accelerator FRAMES demo (see hp).
  • By the 27 Jul call the AMD machines had not yet been received, while HP had shipped the required laptop (see meetings).
  • The project runs on AMD hardware plus google-cloud and GDrive for storage and processing (10 Aug call).
  • The kit supports the AI-generation pipeline (ai-agent-pipeline).

Notes

  • The call notes never split AMD vs HP equipment in detail; Rick (HP) is the delivery contact on the calls.
  • See frames, hp, team, one-small-crawl.

BBC

The BBC is a FRAMES project team member. Nick Kennedy of the BBC is a Contributing Author of the AI Ethics Green Paper.

Identity

  • British public service broadcaster.
  • FRAMES project team member (Contributing Authors).

Role in FRAMES

People

EBU

EBU (European Broadcasting Union) is the alliance of public service media organizations in Europe, North Africa and the Middle East. It co-proposes and co-champions the FRAMES project that hosts One Small Crawl.

Facts

  • full name: European Broadcasting Union (Union Europeenne de Radio-Television)
  • type: alliance / professional association of public service media
  • founded: 1950
  • HQ: Geneva, Switzerland
  • members: ~100+ public service broadcasters (e.g. RAI, BBC, ITV, RTBF, NPO)
  • activity: Eurovision and Euroradio networks, media technology standards (EBU Tech), innovation programmes such as the IBC Accelerator

Role in FRAMES and One Small Crawl

  • FRAMES project Champion and co-proposer (with rai and movielabs). One Small Crawl is an IBC Accelerator 2026 proof-of-concept inside the FRAMES consortium (one-small-crawl).
  • No direct archive material in this production (that role belongs to RAI); EBU provides consortium governance, industry context and showcase reach.

People on One Small Crawl

  • paola-sunna - EBU-side point person on the calls (Eurovision network email domain); artist with supervisor role; IBC 2026 showcase support presenter; maintains shared team learnings and reviews/tags archive material (see meetings).

References (web)

  • show.ibc.org - “FRAMES: Federated Retrieval, Agentic Media Environment…”
  • IBC2026 Accelerator press release (champion list)
  • ebu.ch corporate information

Eddie AI

Eddie AI is a FRAMES Participant and the agentic AI video-editing product evaluated on the calls for one-small-crawl. The June calls name the partner “Eddie AI” / “ADI Assistant Video Editor”; from August the calls call it “Eddie”/“Eddie AI”. Its representative on the calls is shamir.

Tool workflow (as demonstrated on the calls)

  • Ingests footage, scripts and background documents from local or cloud (Google Drive) sources; generates stringouts and rough cuts.
  • Orchestrates multiple AI models (OpenAI, Anthropic, gemini) per task; supports fine-tuning per customer and is developing on-premise deployment.
  • Exports edits to NLEs (resolve, via a separate resolve Cloud project so AI edits do not risk the main project); C2PA compliance planned; prompt conversations exportable for transparency.

Role in the proof-of-concept

  • Proposed for automated rough cuts (8 Jun); adopted for a head-to-head test: Eddie’s automatic edit compared against the manual edit by paola-sunna and roberto-iacoviello (3 Aug, 17 Aug).
  • Paola and Roberto send selected archive footage + script + notes via Google Drive to shamir for the test.

Notes

  • Identity beyond the FRAMES participant listing (legal entity, website) is not confirmed from the sources; the tool and its representative are known from the calls only.
  • See frames, meetings, one-small-crawl.

Eddie AI - agentic AI video editor

Enigmatic

Enigmatic is a FRAMES project team member. Nicolas Frei of Enigmatic is a Contributing Author of the AI Ethics Green Paper.

Identity

  • FRAMES project team member (Contributing Authors).
  • Distinct from eddie-ai (the agentic AI editor, aka ADI AI) - Enigmatic is a separate organisation.

Role in FRAMES

People

Enigmatic - AI media studio (no Wikipedia article; official site)

ETC/USC

ETC/USC (Entertainment Technology Center at the University of Southern California) is a FRAMES project team member. Erik Weaver of ETC/USC is a Contributing Author of the AI Ethics Green Paper.

Identity

  • Full name: Entertainment Technology Center at University of Southern California.
  • FRAMES project team member (Contributing Authors).

Role in FRAMES

People

ETC at USC School of Cinematic Arts (no Wikipedia article; official site)

FRAMES

FRAMES (Federated Retrieval, Agentic Media Environment and Software Defined Workflows) is an IBC (ibc) 2026 Accelerator project: an end-to-end intelligent media production framework that connects broadcaster archives, creative teams and AI agents. One Small Crawl (one-small-crawl) is a proof-of-concept / showcase video produced inside this project.

Overview

FRAMES aims to prove an end-to-end intelligent media production framework that dissolves traditional silos by standardising archival asset management and their metadata and enforcing Zero Trust security. Captured and archive material is transformed in a structured, interoperable environment where Agentic AI can automate tagging, enhancement, and semantic retrieval. Ontology for Media Creation-driven agents work alongside creators who stay in-the-loop to ensure authenticity while remixing archive and newly captured content (omc-ontology, ai-agent-pipeline).

Core challenge objective

Use a knowledge graph and a common semantic model to increase visibility into the production process, allowing clearer and better tradeoffs between creativity, time, and budget.

Approach

  • Starts from a script produced by the team; MovieLabs OMC is used to breakdown scripts and expose narrative and production information to downstream applications.
  • Archive content is tagged per a pre-defined filmmaking dictionary (camera shots and angles, look and feel, narrative).
  • A swarm of locally-running agents performs post-production operations (upscaling, in-painting/out-painting, generative extension, noise removal, audio cleanup) to enhance retrieved content.
  • Event-driven collaboration: a rejected capture makes agents notify sound, colour and VFX specialists with precise improvement instructions (the POC includes a simulated late script-break change).
  • Hybrid capture-and-archive system; the POC delivers a short video as the final project deliverable.

Consortium

  • Champions: EBU (ebu), ITV, ETC/USC, XRBB, BBC, RAI (rai), MovieLabs (movielabs), INSA National Institute of Applied Science (insa, INSA Hauts-de-France Valenciennes), The Global Creative and Security Community.
  • Participants: AMD (amd)/HP (hp), Enigmatic, Streamline Media (streamline-media), eddie-ai, AIMICI. Supporting participant per the IBC2026 press release: Holli (holli).
  • Proposed by RAI, EBU and MovieLabs; kickstart day pitch at IBC2026.

Relationship to One Small Crawl

  • One Small Crawl (production id IBC2026-SNAIL-001) is the animated short made as the FRAMES showcase/POC deliverable. Its archive-plus-AI hybrid pipeline and agent-orchestrated workflow mirror the FRAMES architecture: ARC retrieval agent AGT-016 (federated retrieval from RAI archive), asset_origin tagging, era-locked look applied in post, and the OMC export ingested into this wiki (per-shot OMC JSON, agent registry, OTIO). See one-small-crawl, archive-footage-reuse, animatic-workflow.

References (web)

  • show.ibc.org - “Accelerator Project 2026: FRAMES: Federated Retrieval, Agentic Media Environment and Software (Defined Workflows)” and the IBC2026 Accelerator press release

FRAMES - IBC2026 Accelerator project (no Wikipedia article; IBC official page)

GC-SC

GC-SC (The Global Creative & Security Community) is a FRAMES project team member. Ben Schofield and Aaron Bhugobaun of GC-SC are Project Team First Authors of the AI Ethics Green Paper.

Identity

  • Full name: The Global Creative & Security Community.
  • FRAMES project team member (Project Team - First Authors).

Role in FRAMES

  • Ben Schofield (ben-schofield) and Aaron Bhugobaun (aaron-bhugobaun) are First Authors of the AI Ethics Green Paper (ai-ethics-paper).
  • GC-SC contributes best-practice guidance on measurement, review and continuous improvement (per the paper’s references).

People

GC-SC - Global Creative & Security Community (no Wikipedia article; official site)

HOLLI

HOLLI (holli.st) is a French consultancy for media and entertainment organizations, focused on sustainability, data/AI and organizational change. It is a supporting Participant in the FRAMES project that hosts One Small Crawl.

Facts

  • type: consultancy / technology company (media and entertainment)
  • based: France (clients in cinema, animation, TV, VFX, digital)
  • founders: cedric-lejeune (Workflowers, Ymagis/Eclair, EclairColor) and Franck Isabel; 20+ years combined media-technology experience
  • services: sustainability (CSR integration), organization change, data/AI
  • technology:
    • PRODO - data-driven tool assessing energy and GHG impact of audiovisual processes (supported by CNC, the French film center)
    • HOLLM - private AI processing environment (on-premise/cloud); also the AI production assistant for the mediadata project at insa that manages the One Small Crawl Production Brain (hollm)

Role in FRAMES and One Small Crawl

  • Supporting Participant of FRAMES per the IBC2026 press release (“supporting Participants AMD-HP and Holli”). Co-championed the related 2026 pitch “Software Defined Workflows for Interoperable Movie and TV Production” with movielabs.
  • One Small Crawl is the IBC Accelerator 2026 proof-of-concept inside FRAMES (one-small-crawl).
  • Likely relevance: the bible’s sustainability constraint SC-008 requires agents to log GPU hours, energy consumption and estimated CO2 per generation task (AGT-017 reporting) - directly aligned with HOLLI’s PRODO sustainability focus and the film’s AI-generation pipeline (ai-agent-pipeline).

People on One Small Crawl

  • cedric-lejeune - HOLLI co-founder; on the production as admin/supervisor (cl@holli.st account). Runs the local kitsu production infrastructure and resolve post/colour work; assembly of animatics + grades with chris-vienneau (see meetings).
  • Franck Isabel - HOLLI co-founder (with Cedric); not part of the One Small Crawl crew (no page yet).

References (web)

  • show.ibc.org - FRAMES project page and 2026 Accelerator press release
  • holli.st corporate site (mission, services, PRODO)

HOLLI - HOLLI/mediadata (no Wikipedia article; official site)

HP

HP (Hewlett Packard) is a US technology company. With AMD it is listed as the FRAMES Participant “AMD/HP” supplying hardware for the frames proof-of-concept one-small-crawl, and HP is also an IBC2026 sponsor.

Role in FRAMES and One Small Crawl

  • Supplies the production workstations: HP Z-series machines were discussed from the 7 Apr call onward (Erik shared a NetBird + Create reference architecture for them).
  • Provides the HP Z2 mini workstation that powers hollm for the IBC Accelerator FRAMES demo (with the AMD Strix Halo processor, see amd).
  • Kit delivery is tracked with Rick (HP) - the notes transcribe his surname as “Holman” (13 Jul) and “Pullman” (22 Jun) - and the crew’s AMD/HP hardware order is coordinated by roberto-iacoviello (see meetings).
  • On 27 Jul HP had shipped the required laptop while the AMD machines were still pending.
  • IBC2026 marketing lists HP among the event sponsors (Google/HP/Shaw).

Notes

IBC

IBC (International Broadcasting Convention) is the organization behind the annual IBC trade show and the IBC Accelerator programme that hosts the FRAMES project (frames) and One Small Crawl (one-small-crawl). Its web presence is show.ibc.org.

Facts

  • full name: International Broadcasting Convention
  • type: trade association / event and media-innovation organizer
  • activity: annual IBC show (Amsterdam), IBC Accelerator media innovation programme, awards, conferences
  • relevant edition: IBC2026 - the Accelerator cohort that runs FRAMES

Role in FRAMES and One Small Crawl

  • Runs the IBC Accelerator programme; FRAMES is an IBC 2026 Accelerator project (frames), and One Small Crawl is its showcase/POC deliverable (one-small-crawl, production id IBC2026-SNAIL-001).
  • Sets Accelerator logistics: showcase pods, MPTS spotlight, contracts, project-page publication, deadlines (deck, sizzle, ethics paper), and the official note-taker policy seen on the weekly calls (meetings).

People on One Small Crawl

  • muki - IBC staff: coordinator for the FRAMES Accelerator project on the weekly calls (contracts, logistics, event coordination).

References (web)

INSA

INSA (Institut National de Sciences Appliquees / National Institute of Applied Sciences) is the French network of public engineering “grandes ecoles”. It is listed as a Champion in the FRAMES project that hosts One Small Crawl.

Facts

  • full name: Institut National de Sciences Appliquees (INSA; accents folded to ASCII per wiki policy)
  • type: public engineering and research institute (network of “grandes ecoles” under the French Ministry of Higher Education)
  • founded: 1957 (first school, INSA Lyon)
  • campuses: Lyon, Rennes, Rouen, Strasbourg, Toulouse, Centre Val de Loire, Hauts-de-France (plus INSA Euro-Mediterranee in Morocco)
  • research strengths: engineering, computer science, AI, image/audio processing, XR and immersive media

Role in FRAMES and One Small Crawl

  • Listed as a FRAMES Champion on the IBC2026 project page (“INSA National Institute of Applied Science”), alongside rai, ebu and movielabs. One Small Crawl is the IBC Accelerator 2026 proof-of-concept inside FRAMES (one-small-crawl).
  • The FRAMES participant is INSA Hauts-de-France in Valenciennes (confirmed 2026-09-02 by the production team). INSA Hauts-de-France (Campus Mont-Houy, Valenciennes) is the newest INSA school (est. 2019); its Audiovisual and Multimedia engineering programme teaches XR production, spatial computing and generative AI - aligned with the FRAMES media/AI focus.
  • Expected role: research/academic Champion input to archive tagging, semantics and agentic workflows in the proof-of-concept.

References (web)

ITV

ITV is a FRAMES project team member. Clive Santamaria of ITV is a Contributing Author of the AI Ethics Green Paper.

Identity

  • British commercial television broadcaster.
  • FRAMES project team member (Contributing Authors).

Role in FRAMES

People

MedIAdata

MedIAdata is a partnership between INSA Hauts-de-France and HOLLI to develop agentic production and run experiments connecting research, teaching, industry, and solution providers in the cultural and creative industries (ICC). It is the institutional vehicle that supports HOLLM and the team’s participation in FRAMES / One Small Crawl.

  • Legal form: French association loi 1901 (non-profit)
  • Based: Hauts-de-France, France
  • Mission: an experimentation space around data, formats, and new modes of production and diffusion for cultural and creative industries
  • Website: www.mediadata.tech (2026)
  • Contact: contact@mediadata.tech

Role in FRAMES and One Small Crawl

INSA Hauts-de-France participates in the FRAMES project via MedIAdata (2026-06-25-frames-call): the team’s research and agentic workflow contributions are formalized through the MedIAdata association.

MedIAdata funds and staffs the development of HOLLM, the AI production assistant that manages the One Small Crawl Production Brain (kitsu + graphdb + Obsidian vault pipeline).

cedric-lejeune is the tech manager representing MedIAdata in FRAMES (HOLLI co-founder).

Featured in the IBC2026 Accelerator Zone: “MedIAdata sera present a l’IBC grace a la participation de l’INSA Hauts-de-France au projet FRAMES” (per the official site).

Four pillars (poles)

  1. Production - data continuity from design to distribution: pipelines, metadata, tool interoperability.
  2. Diffusion - formats and platforms, new channels (XR, metaverse, broadcast in virtual worlds).
  3. Recherche (Research) - reconnecting academic work with field practice; participation in international standards bodies.
  4. Formation (Training) - shared pedagogical resources, technology watch, skills transformation.

Use cases (cas d’usage)

Three 6-month experimental projects, each run collaboratively between clients, vendors, students, and researchers:

  • Animation - data continuity from production to distribution, with automated agent-based monitoring.
  • Fiction - shoot, post-production, delivery: documenting and tooling an entire chain end-to-end.
  • XR & live - integrating interactivity into new formats.

Governance (per the official site)

  • tech: cedric-lejeune - production technology expert, founder of holli
  • recherche: Renan Mouren - Maitre de conferences, insa Hauts-de-France / UPHF
  • studios & production: Dominique Joutaud - animation producer

Notes

  • MedIAdata is an association law 1901 in Hauts-de-France; supported by INSA Hauts-de-France and by HOLLI.
  • The MedIAdata project is the institutional umbrella under which HOLLM (the AI production assistant) and the One Small Crawl Brain tooling sit.
  • See holli (HOLLI’s role), insa (academic partner), cedric-lejeune (tech manager), hollm (the deliverable), frames (the IBC showcase), one-small-crawl (the proof of concept).
  • Web link: www.mediadata.tech – 2026.

MovieLabs

MovieLabs (Motion Picture Laboratories, Inc.) is the technology research joint venture of the major Hollywood studios. It is the author of the OMC (Ontology for Media Creation), the data model that governs this production.

Facts

  • full name: Motion Picture Laboratories, Inc.
  • type: not-for-profit technology research and development organization
  • founded: ~2006 (Palo Alto area, California, USA)
  • ownership: joint R&D venture of the major motion picture studios (Paramount, Sony, Universal, Disney, Warner Bros.)
  • flagship work: OMC (Ontology for Media Creation), the 2030 Vision, software-defined workflows

Role in FRAMES and One Small Crawl

  • FRAMES project Champion and co-proposer (with rai and ebu). One Small Crawl is an IBC Accelerator 2026 proof-of-concept inside FRAMES (one-small-crawl).
  • Author of the ontology this production runs on: the Agent Production Bible states OMC was developed by MovieLabs with DreamWorks Animation, Marvel Studios, Paramount, Sony Pictures, Universal, Walt Disney Studios and Warner Bros. (omc-ontology).
  • The authoritative OMC export ingested into this wiki (per-shot OMC JSON, agent registry, naming schema, OTIO timeline) is a MovieLabs-format deliverable (naming-schema, ai-agent-pipeline, animatic-workflow).

References (web)

  • show.ibc.org - “FRAMES: Federated Retrieval, Agentic Media Environment…”
  • IBC2026 Accelerator press release (champion list)
  • movielabs.com - OMC and 2030 Vision publications

Press Play Labs

Press Play Labs is the organisation of Shamir Alibhai, a Contributing Author of the AI Ethics Green Paper.

Identity

  • FRAMES project team member (Contributing Authors).

Role in FRAMES

People

  • shamir - Contributing Author (Press Play Labs)

Press Play Labs - Shamir Alibhai’s company (no Wikipedia article; official site)

RAI

RAI (Radiotelevisione Italiana S.p.A.) is the Italian national public broadcasting service and one of Europe’s largest media organizations. Its archive is a primary content source for One Small Crawl.

Facts

  • full name: Radiotelevisione Italiana (Italian Radio-Television)
  • type: state-owned public service broadcaster
  • founded: 1954 (successor to URI/EIAR radio services dating from 1924)
  • HQ: Rome, Italy
  • services: national TV networks (Rai 1, Rai 2, Rai 3), radio, news, docs
  • archive: one of Europe’s largest audiovisual collections (news, docs, entertainment); RAI Teche division

Role in FRAMES and One Small Crawl

Crew on One Small Crawl (RAI)

Four of the human crew are RAI staff (rai.it accounts; see team):

  • roberto-iacoviello - supervisor; named lead AI artist in the call role list; cross-cutting coordinator owning the most action items; leads archive-metadata collection, rights and script review; opens the IBC 2026 showcase presentation.
  • gabriele-monteleone - supervisor; named DP (director of photography); owns 3D/unreal blocking and character assets. Matches the on-disk collaboration alias “gabri” (RAI OneDrive tree).
  • alberto-ciprian - artist (supervisor role); snail character design, loraing and the era environments; develops the snail + archive compositing workflow.
  • davide-zappia - supervisor; runs the archive-metadata extraction workflow (frame-level vs sampled analysis).

References (web)

  • show.ibc.org - “FRAMES: Federated Retrieval, Agentic Media Environment…”
  • IBC2026 Accelerator press release (champion list)
  • rai.it corporate information

Streamline Media

Streamline Media is listed as a Participant (technology/supplier side) in the FRAMES project at IBC2026, the consortium that hosts One Small Crawl.

Identity

  • Web presence: streamline-media.net (production platform), with the One Small Crawl project work area hosted at app.streamline-media.net/project/13b69d06-e315-4d23-a1df-61821b339c5e/work/
    • that app URL is the company’s own platform (per john-hobbs).
  • The FRAMES project page lists the Participant name “Streamline Media”; the in-production corroboration is John Hobbs’ streamline-media.net account (john@streamline-media.net), admin/supervisor on One Small Crawl.
  • NOT the unrelated companies that share the name (e.g. streamline-media.com web agency, or Streamline Media Group / streamline-mediagroup.com - a different holding company).

Role in FRAMES and One Small Crawl

  • Listed as a FRAMES Participant alongside AMD (amd)/HP (hp), Enigmatic, eddie-ai and AIMICI. Participants are technology vendors that build and execute the proof-of-concept with the Champions (rai, ebu, movielabs).
  • John Hobbs’ company (streamline-media.net) is the Participant’s operational face on the production: John is admin/supervisor, composes the score (developed with suno), and hosts production assets for review on the Streamline platform (unreal-render-pipeline connects only insofar as the platform hosts review copies; no unreal-specific claim is made).
  • Open question: the exact legal-entity name in FRAMES is still not spelled out beyond “Streamline Media”, but its web presence is confirmed as streamline-media.net.

People on One Small Crawl

  • john-hobbs - admin/supervisor on the production (john@streamline-media.net); composes the film’s score (developed with suno) and hosts production assets for review on the Streamline platform. His streamline-media.net account is the in-production corroboration for this entity (see Identity above).

References (web)

Streamline Media - John Hobbs’ company (no Wikipedia article; official site)

XRBB

XRBB (Extended Reality Berlin-Brandenburg) is a FRAMES project team member. Tim Deussen of XRBB is a Project Team First Author of the AI Ethics Green Paper and wrote the paper’s closing word.

Identity

  • Full name: Extended Reality Berlin-Brandenburg.
  • FRAMES project team member (Project Team - First Authors).

Role in FRAMES

  • Tim Deussen (tim-deussen) is a First Author of the AI Ethics Green Paper (ai-ethics-paper) and contributed the closing argument on the “proof” problem in AI-assisted production.

People

XRBB - Extended Reality Berlin-Brandenburg (no Wikipedia article; official site)

Yamdu

Cloud-based production management software for film and media productions. On One Small Crawl, florian-reimann represents Yamdu on the FRAMES team – the call notes show him attending most 2026-07 through 2026-08 calls.

  • Website: yamdu.com
  • Type: production management SaaS
  • Position on One Small Crawl: tool / vendor rep – production-management perspective, call-sheet / scheduling workflow

What it does

  • Script breakdowns (elements extraction, scene tagging)
  • Shooting schedules (stripboards, day-out-of-days)
  • Call sheets (crew / cast dispatch, location, weather)
  • Time tracking, billing, cost reporting
  • Cloud-based collaboration across departments (production, AD, locations, transport)

On One Small Crawl

Yamdu - cloud production-management SaaS (no Wikipedia article; official site)

People

  • Aaron Bhugobaun
  • Ahsan Mallick
  • Aidan Enigmatic
  • Alberto Ciprian
  • Ben Schofield
  • Cedric Lejeune
  • Chris Vienneau
  • Clive Santamaria
  • Davide Zappia
  • Erik Weaver
  • Florian Reimann
  • Gabriele Monteleone
  • John Canning
  • John Hobbs
  • Kathryn Webb
  • Muki Kulhan
  • Nick Kennedy
  • Nicolas Frei
  • Paola Sunna
  • Raymond Drewry
  • Roberto Iacoviello
  • Shamir Allibhai
  • Team and Agents: humans (11 kitsu accounts on the production) + AI agents (AGT-xxx)
  • Tim Deussen

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

Aaron Bhugobaun

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). Chapter Forge Media account (chapterforgemedia.com email domain), position supervisor.

kitsu account

  • email: aaron@chapterforgemedia.com
  • country: GB (United Kingdom)
  • role: supervisor
  • position: supervisor
  • active: true

Notes

  • Not referenced in the 17 FRAMES weekly call notes ingested under raw/meetings/ (meetings); no call-level activity recorded as of 2026-08-17-frames-call.
  • chapterforgemedia.com domain suggests an affiliation with Chapter Forge Media (a media/VFX production company in the UK); no organisation page yet.
  • See one-small-crawl, team.

Ahsan Mallick

Ahsan Mallick (AIMICI) is a Project Team First Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

  • Project Team - First Author of the AI Ethics Green Paper, representing aimici.

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

Aidan Enigmatic

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). Enigmatic account (enigmatic.swiss email domain), position artist with user role.

kitsu account

  • email: aidan@enigmatic.swiss
  • country: CH (Switzerland)
  • role: user
  • position: artist
  • active: true

FRAMES weekly calls (2026)

Notes

  • enigmatic.swiss domain suggests an affiliation with Enigmatic (FRAMES Participant per the IBC2026 press release); no organisation page yet.
  • See one-small-crawl, team.

Alberto Ciprian

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). RAI account (rai.it email domain), position artist with supervisor role.

kitsu account

  • email: alberto.ciprian@rai.it
  • role: supervisor
  • position: artist
  • active: true

FRAMES weekly calls (2026)

Notes

Ben Schofield

Ben Schofield (GC-SC) is a Project Team First Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

  • Project Team - First Author of the AI Ethics Green Paper, representing gc-sc (The Global Creative & Security Community).
  • GC-SC contributes best-practice guidance on measurement, review and continuous improvement (per the paper’s references).

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

Cedric Lejeune

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). HOLLI co-founder (with Franck Isabel) and admin on the production. This page documents the cl@holli.st account; “Super Admin” is the role/position label for this person on kitsu (instance admin), not a separate account.

kitsu account

  • email: cl@holli.st
  • country: FR (France)
  • role: admin
  • position: supervisor (instance “Super Admin” on kitsu)
  • active: true

FRAMES weekly calls (2026)

Notes

  • HOLLI affiliation links to holli (FRAMES supporting participant, sustainability/Data-AI consultancy).
  • The on-disk collaboration tree “ce” (Videos/Projects/OneSmallCrawl, CedricTests; editor/VFX experiments) matches this person (team).
  • Represents the mediadata project in FRAMES: mediadata is a partnership between holli and insa, for which Cedric is the tech manager (reported by user, 2026-09-02).
  • See one-small-crawl.

Chris Vienneau

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). Part of the human crew; position lead with supervisor role on the production.

kitsu account

  • email: chris@fx-dmz.com
  • country: CA (Canada)
  • role: supervisor
  • position: lead
  • active: true

FRAMES weekly calls (2026)

Notes

  • The on-disk collaboration tree “chris” (Videos/Screen Recordings; screen recordings for animatics/edits) matches this kitsu account (team).
  • See one-small-crawl.

Clive Santamaria

Clive Santamaria (ITV) is a Contributing Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

  • Project Team - Contributing Author of the AI Ethics Green Paper, representing itv.

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

Davide Zappia

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). RAI account (rai.it email domain), position supervisor.

kitsu account

  • email: davide.zappia@rai.it
  • country: IT (Italy)
  • role: supervisor
  • position: supervisor
  • active: true

FRAMES weekly calls (2026)

Notes

Erik Weaver

Erik Weaver (ETC/USC) is a Contributing Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

  • Project Team - Contributing Author of the AI Ethics Green Paper, representing etc-usc (Entertainment Technology Center at USC).

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

Florian Reimann

Florian Reimann represents yamdu on the One Small Crawl FRAMES team. Cloud-based production management software; the Yamdu angle is script breakdowns, call sheets and time tracking.

Role

  • FRAMES Team - representing yamdu (production-management vendor).
  • External contact (no kitsu account on the One Small Crawl production).
  • Participated in the FRAMES weekly calls from 2026-07-20-frames-call onward (meetings).

Notes

Gabriele Monteleone

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). RAI account (rai.it email domain), position supervisor.

kitsu account

  • email: gabriele.monteleone@rai.it
  • country: IT (Italy)
  • role: supervisor
  • position: supervisor
  • active: true

FRAMES weekly calls (2026)

Notes

  • RAI affiliation links to rai.
  • The on-disk collaboration tree “gabri” (RAI OneDrive tree; likely RAI liaison/archive contact) matches this kitsu account (team).
  • See one-small-crawl.

John Canning

John Canning (AMD) is a Contributing Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

  • Project Team - Contributing Author of the AI Ethics Green Paper, representing amd.

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

John Hobbs

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). Streamline-Media account (streamline-media.net email domain), admin role with supervisor position.

kitsu account

  • email: john@streamline-media.net
  • country: GB (United Kingdom)
  • role: admin
  • position: supervisor
  • active: true

FRAMES weekly calls (2026)

Notes

John Hobbs - Streamline Media co-founder

Kathryn Webb

Kathryn Webb (AIMICI) is a Project Team First Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

Muki Kulhan

Muki Kulhan works for IBC (ibc), the organization that runs the IBC Accelerator programme, as coordinator of the FRAMES IBC2026 Accelerator project that hosts one-small-crawl. On the 30 Mar call Muki is listed with the role “IBC coordinator”. Not part of the creative crew and not a kitsu account on the production (team).

Role on the calls

  • Project logistics: IBC Accelerator contracts and tier levels, company logos (to Octavia), slack onboarding, project-page publication, official IBC-note-taker-only policy.
  • Programme/event coordination: MPTS spotlight, IBC showcase pod rota and showcase-slot booking, deadlines (deck, sizzle, ethics paper).
  • Administrative owner/co-owner of many action items (contracts, access requests, google-cloud nudges, ethics-paper drafting with Kathryn).

Notes

Nick Kennedy

Nick Kennedy (BBC) is a Contributing Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

  • Project Team - Contributing Author of the AI Ethics Green Paper, representing bbc.

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

Nicolas Frei

Nicolas Frei (Enigmatic) is a Contributing Author of the AI Ethics Green Paper (ai-ethics-paper).

Role

  • Project Team - Contributing Author of the AI Ethics Green Paper, representing enigmatic.

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.

Paola Sunna

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). EBU-affiliated account (eurovision.net email domain), position artist with supervisor role.

kitsu account

  • email: sunna@eurovision.net
  • country: IT (Italy)
  • role: supervisor
  • position: artist
  • active: true

FRAMES weekly calls (2026)

Notes

  • Eurovision email domain (Eurovision network is EBU’s operational arm) links to ebu, FRAMES Champion and consortium governance for the production.
  • See one-small-crawl, team.

Raymond Drewry

Raymond Drewry (MovieLabs) is a Project Co-Lead and First Author of the AI Ethics Green Paper (ai-ethics-paper). Email: rdrewry@movielabs.com.

Role

  • Project Co-Lead & First Author of the AI Ethics Green Paper, representing movielabs.

Notes

  • External paper author; not a kitsu account on the One Small Crawl production.
  • Surname spelled “Drewry” (user-confirmed); the ethics paper source lists “Raymond Drury” - treated as the same person.

Roberto Iacoviello

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). RAI account (rai.it email domain), position supervisor.

kitsu account

  • email: roberto.iacoviello@rai.it
  • role: supervisor
  • position: supervisor
  • active: true

FRAMES weekly calls (2026)

Notes

Shamir Allibhai

Shamir Allibhai is the eddie-ai representative on the FRAMES calls. A new participant introduced on the 29 Jun call (with Kathryn from Amici). Not part of the creative crew and not a kitsu account on the production (team).

Role on the calls

  • eddie-ai point of contact: paola-sunna and roberto-iacoviello send selected archive footage + script + notes via Google Drive to Shamir for the eddie-ai rough-cut test (3 Aug; 17 Aug).
  • Demonstrated eddie-ai - ingest, stringouts/rough cuts, model orchestration, NLE export, C2PA plans - on the 3 Aug and 17 Aug calls.
  • C2PA support: offered implementation assistance and community contacts (29 Jun); discusses exporting Eddie prompt conversations with Kathryn (17 Aug) for transparency.

Notes

Team and Agents

The production crew of One Small Crawl (human) plus the registered AI agents (AGT-xxx) that operate under the Agent Production Bible. The human crew below is authoritative from the local kitsu instance (localhost:5000): the 11 accounts registered on the One Small Crawl production.

Humans (kitsu accounts on One Small Crawl production)

  • chris-vienneau - Chris Vienneau - chris@fx-dmz.com - supervisor, lead
  • davide-zappia - Davide Zappia - davide.zappia@rai.it - supervisor
  • paola-sunna - Paola Sunna - sunna@eurovision.net - supervisor, artist
  • cedric-lejeune - Cedric Lejeune - cl@holli.st - admin, supervisor
  • roberto-iacoviello - Roberto Iacoviello - roberto.iacoviello@rai.it - supervisor
  • gabriele-monteleone - Gabriele Monteleone - gabriele.monteleone@rai.it - supervisor
  • tim-deussen - Tim Deussen - tim.deussen@gmail.com - user, artist
  • john-hobbs - John Hobbs - john@streamline-media.net - admin, supervisor
  • aidan-enigmatic - Aidan Enigmatic - aidan@enigmatic.swiss - user, artist
  • alberto-ciprian - Alberto Ciprian - alberto.ciprian@rai.it - supervisor, artist
  • aaron-bhugobaun - Aaron Bhugobaun - aaron@chapterforgemedia.com - supervisor

On-disk collaboration tree (earlier, non-authoritative)

  • gabri - RAI OneDrive tree; likely RAI liaison/archive contact (matches kitsu: Gabriele Monteleone @ RAI)
  • ce - Videos/Projects/OneSmallCrawl, CedricTests; editor/VFX experiments (matches kitsu: Cedric Lejeune)
  • chris - Videos/Screen Recordings; screen recordings for animatics/edits (matches kitsu: Chris Vienneau)
  • AI Supervisor - maintains the bible (per document control)
  • Orchestrator Agent (human-authorized) - coordinates agents

AI agents (registered in agent-registry)

  • AGT-002 Gen-Image / AGT-003 Gen-Video - character animation, VFX, plates
  • AGT-004 Gen-Audio - TTS, SFX, voice with SAG-AFTRA consent
  • AGT-011 Color Pipeline - era-matches color between archive and AI layers
  • AGT-016 ARC Retrieval - searches archives for era-authentic footage
  • AGT-019 Enhancement - upscales, stabilizes, inpaints
  • AGT-020 Shot-List Agent - produced the 45-shot list (gemini-3.1-pro-preview)
  • Script-Analysis, ASR/Transcription, Translation, AI Metadata, QC, Scheduling, Report-Gen agents

Governance

  • All agents MUST follow the Agent Production Bible; bible_version checked at session and task start; stale versions HALT.
  • Every asset tagged with asset_origin (archive vs ai_generated vs ai_enhanced vs ai_composite).
  • Approvals via approval token pipeline; batch review (AI dailies).

Notes

Tim Deussen

Team member of One Small Crawl, registered on the production in kitsu (localhost:5000). Position artist with user role; affiliation not stated in kitsu (personal gmail domain).

kitsu account

  • email: tim.deussen@gmail.com
  • role: user
  • position: artist
  • active: true

FRAMES weekly calls (2026)

  • Listed among participants on all 17 FRAMES weekly call notes in raw/meetings/ (meetings), 30 Mar - 17 Aug. Shot-list / data-side backbone of the project: owns the master Excel shot list and much of the archive metadata workflow.
  • Key call activity (per-call pages under meetings):
  • Sources: the 17 FRAMES weekly call notes under raw/meetings/ (see meetings).

Notes

  • No organisation page yet: affiliation not declared in kitsu. If the affiliation is confirmed (e.g. FRAMES consortium partner), add an organisation link here.
  • See one-small-crawl, team.

Production

  • One Small Crawl: the film: IBC2026-SNAIL-001, animated short, 5 eras,
  • Archives: 1 sub-items
  • Characters: 1 sub-items
  • Props
  • Environments: 1 sub-items
  • Deliverables: 1 sub-items
  • Sequences: 5 sub-items

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

One Small Crawl

Animated short film built for the IBC Accelerator 2026 showcase, as a proof-of-concept video inside the FRAMES project (frames). Production id IBC2026-SNAIL-001, working title One Small Crawl. Hybrid production: authentic archival footage retrieved via the ARC pipeline plus AI-generated character animation, VFX, era-style enhancement, and sound design. NOT fully AI-generated - provenance of each layer is tracked.

Key facts

  • format: animated_short; genre animation/drama
  • runtime_target: 5-8 minutes; delivery 4K UHD (ACES AP1 working, Rec2020)
  • scene_count: 5 - one era of film/video history per scene
  • logline: a snail (snail) walks the full length of route-66 from Chicago to Santa Monica across decades of American media history
  • characters: CHAR-001 Snail (protagonist, minimal dialogue - 4 short lines, quiet, private, awed); CHAR-002 Voice From TV (NASA “the Eagle has landed”, archival audio, clearance to verify)
  • product_placement: Nutella jar floating zero-g in lunar orbit capsule (sc04), no dialogue; clearance PENDING
  • delivery: IBC Accelerator 2026 showcase + festival circuit

Scenes (era arc)

  • sc01 “The Road Begins” - 1926-window Chicago, silent film (sf)
  • sc02 “Dust and Motion” - era-locked-look Dust Bowl, sound newsreel (snd), Ariston Cafe Litchfield IL
  • sc03 “The World Turns to Color” - July 20 1969 (Apollo 11 landing day), Kodachrome pivot, U-Drop Inn Shamrock TX (kch)
  • sc04 “Somebody’s Tape” - mid-1980s AZ-CA desert, consumer VHS (vhs), incl. lunar capsule + Nutella zero-g shots (sp/hd sub-eras)
  • sc05 “Arrival” - 2020s Santa Monica, modern digital (dig), golden hour

The same 5-scene structure is tracked in the local kitsu instance as sequences sc01-sc05 (sc01 … sc05), each broken into shots (sc01_0010 … sc05_0060, 50 records, 18 canceled). kitsu sequence titles differ from the scene names above (e.g. sc01 “The Route 66 Sign”). Each shot page is enriched with the movielabs OMC export JSON: generation brief v0.3.0 (50/50 shots), shot list v0.2.0 revised with AGT-020 fallback, and the per-shot OMC sidecar (34/50) recording model, seed, era, and deliverable stem.

Timeline

  • 2026-05-16: bible v1 effective; script v2 draft date
  • 2026-06-07: shot list produced by AGT-020
  • 2026-06-28: script draft v2.2.0
  • 2026-07-04/05: era look guide research-locked; resolve era stacks built
  • 2026-09-02: MovieLabs OMC export received and ingested into this wiki

Relationship to archive assets on disk

  • The film reuses archive home-movie and news footage (archive-footage-reuse): “Our Trip of U.S.A. Part Two” (6403-our-trip-of-usa) and RAI news records (Route 66 + Artemis II, see artemis-ii-tg3-news).
  • The “one small crawl” pun plays on “one small step” - a snail’s crawl vs spaceflight (artemis-ii theme in the TV voice and capsule scenes). Confirmed by the script and era guide, not just inferred.

Pipeline

Open questions

  • Exact runtime vs 5-8 min target.
  • Which RAI archive items made the cut (Route 66 docs, Artemis II, Harley, Kilimangiaro, TG specials are all candidates in the export).
  • Nutella clearance status.

Archives

  • Artemis II: space thread: NASA mission + RAI news + “Eagle has landed”

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

Artemis II

NASA crewed mission around the Moon (Artemis program). In the One Small Crawl production, Artemis II material is space-news archival source - part of the film’s spaceflight thread that plays against the snail’s crawl.

Role in the film

  • The NASA “Voice From TV” (CHAR-002) is “the Eagle has landed” - Apollo 11 mission archival audio, used as the space/moon motif (public domain / archival, clearance to verify).
  • The film’s space tangent culminates in sc04’s lunar-orbit capsule sub-sequence: snail in orbit, Nutella floating zero-g, Earth seen through a porthole (era codes sp = capsule interior, hd = orbital).
  • The working-title pun “one small crawl” / “one small step” pairs the snail’s scale with spaceflight (snail returns from walking route-66 to an orbital track).

Archive on disk / in export

  • RAI Tg3 news records in the OMC export: T09143_211 and T26097_303 “Tg3 Pomeriggio - Missione spaziale Artemis II” (see artemis-ii-tg3-news).
  • Original wiki read the Artemis II clip as a “then vs now” juxtaposition; the authoritative bible shows it as part of the film’s space motif (archive-footage-reuse).

Notes

Characters

  • Snail (Stella): Stella, CHAR-001, the snail protagonist walking Route 66

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

Snail (Stella)

CHAR-001 - the protagonist of One Small Crawl. A cartoon snail with round eyes and an unhurried disposition who walks the full length of route-66 from Chicago to Santa Monica across decades. The story’s “crawl” is hers: the film is a road trip at snail scale through American media history.

Name

  • The FRAMES team agreed the snail is named Stella on the 2026-07-13-frames-call weekly call (raw/meetings/2026-07-13-frames-call-frames-call.md), alongside confirming the film title “One Small Crawl”; confirmed again on 2026-07-20-frames-call (raw/meetings/2026-07-20-frames-call-frames-call.md). This is the authoritative name.
  • The PeriodO crosswalk (raw/movielabs/show-ont-periodo-crosswalk.md) calls the character “Olivia” (medium confidence). That name predates the team’s explicit decision and is superseded by it; kept only as a historical note.

Character design

  • Cartoon snail, round eyes, simple, expressive; consistent across all eras
  • Minimal dialogue: 4 short lines total; quiet, private, awed
  • Generated as a separate layer on transparent alpha (asset_origin: ai_generated), composited over archive backgrounds by a human VFX artist; the snail + archive compositing workflow is developed by alberto-ciprian (see ai-agent-pipeline)

Story arc (per shot list + era guide)

  • sc01: departs from the eastern terminus of Route 66 - Michigan Avenue at the Art Institute of Chicago, beneath the lions; wonder -> determined; a sputtering Model T ignores her
  • sc02: keeps pace with a world in flight during the migration years; determined, undertone sad; pauses at a diner (Ariston Cafe) to listen
  • sc03: color bleeds from a rainbow; wonder (Kodachrome era, July 20 1969)
  • sc04: somebodys tape - content; includes orbital/capsule sub-sequence (snail in a lunar orbit capsule with Nutella floating zero-g)
  • sc05: arrival in Santa Monica, ocean stillness, look back; content + wonder

Notes

Props

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

Environments

  • Route 66: the road: Chicago to Santa Monica, per-scene geography

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

Route 66

The geographic spine of One Small Crawl. The snail walks the full length of the historic US highway Route 66 from Chicago (eastern terminus) to Santa Monica, California (western terminus) - across decades, one film/video era per scene. It is both the literal road and the narrative/metaphoric “road through American media history.”

Narrative geography (per scene)

  • sc01: Chicago - eastern terminus, Michigan Avenue at the Art Institute of Chicago beneath the lions (1926-window, silent)
  • sc02: Midwest Dust Bowl era - Ariston Cafe, Litchfield IL (coords 39.1764,-89.6646 verified); MO-OK-TX newsreel
  • sc03: U-Drop Inn, Shamrock TX - Kodachrome pivot, July 20 1969
  • sc04: AZ-CA desert - consumer VHS mid-1980s; plus a lunar-orbit capsule sub-sequence (space tangent, see artemis-ii)
  • sc05: Santa Monica - modern digital arrival, golden hour, ocean

PeriodO periods

  • p-osc-narr-sc1 (Chicago origin, 1926-1927, deny 1927-1929)
  • p-osc-narr-sc2 (Peak Route 66 Midwest, ~1935-1948, Ariston Cafe)

Research anchoring

  • Real, named, still-operating places depicted as documented historical fact, with per-shot OMC sidecars recording location, source photo/archive, and date depicted.
  • oscx:heroLocations: named real locations with NRHP IDs, bypass dates, archival provenance (LOC Highsmith call numbers, Villanova PDTP archive, Burton Frasher/Pomona).
  • NRHP properties DB: Soulsby, Ariston, U-Drop, Blue Swallow.
  • Route 66 sign evolution asset: five sign generations, fabrication cheat sheet, weathering logic, legal rulings.

Notes

Deliverables

  • Deliverables: the film’s output deliverables (final, YouTube export, score)

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

Deliverables

The output deliverables of one-small-crawl for the IBC Accelerator 2026 showcase and festival circuit.

Primary deliverable

  • Final film - animated short, 4K UHD (ACES AP1 working, Rec2020), runtime target 5-8 minutes, 5 scenes (one era of film/video history each).
  • Delivery: IBC Accelerator 2026 showcase + festival circuit.

Export / render

Process

  • Conform in resolve - the edit is conformed in DaVinci Resolve from the OTIO editorial timeline (see otio-timeline, animatic-workflow).
  • sRGB/linear color pipeline for the effect timelines - effects work is done on timelines in an sRGB/linear color pipeline, so the VFX/compositing layers are graded and matched in a linear working space (see era-locked-look, ves-good-practices).
  • Finish in ACES - the conformed, effect-completed timelines are finished in ACES (ACES AP1 working space, Rec2020 primaries) to deliver both an HDR and an SDR version of the film.

Master QC

The finished HDR and SDR masters are checked before delivery (see ves-good-practices for the color discipline these checks enforce).

  • Color decisions in the screening room - HDR and SDR masters are graded and QC’d in the screening room, not at an artist’s desktop: the master must be signed off under the display environment it is aimed at.
  • Per-version monitoring - the HDR master is viewed/monitored on an HDR display (its mastering display) and the SDR master on an SDR reference monitor; never QC one version’s look on the other’s display.
  • Display transform never baked in - masters are checked in the working space and conformed to the deliverable encoding at export; the scene-referred master keeps the output/display transform out of the graded image.
  • No harm to the plate - QC verifies VFX regions integrate and untouched plate regions are unmodified; round-trip test unmodified pixels match the original plate for the DI.
  • Technical QC - codec/container conformance, resolution, frame rate, bit depth, audio sync/silence/true-peak, and that no illegal values or clipping result from the ACES AP1/Rec2020 finish.
  • Sign-off - the one-small-crawl team + post/color reviewers approve each master (HDR and SDR) before it is logged to the OMC sidecar and delivered.

File naming

  • Deliverable files follow the production-wide naming-schema (OSC v1.0), which extends TR-VFX-IS with asset origin, agent/vendor codes, era codes, and the OMC version model - every deliverable name encodes its full provenance.
  • Per VES/ETC good practice (see ves-good-practices): version by revision (never by shot number), fix the fixed fields first, keep the filename independent of its directory, and use only the allowed character set.
  • Example deliverable stem: osc_sc02_0100_cmp_gen_a003_v003_ntrl_snail-walk.1001.exr (see naming-schema for the full syntax).

Supporting deliverables

Notes

Sequences

  • 01 The Route 66 Sign: 5 sub-items
  • 02 The Shoulder & the Diner: 7 sub-items
  • 03 The Rain & the TV: 10 sub-items
  • 04 The Decay & the Dream: 22 sub-items
  • 05 The End: 6 sub-items

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

sc01 - 01 The Route 66 Sign

Sequence sc01 - The Route 66 Sign

Preview

First available Kitsu preview for this sequence (from shot sc01_0010). 4 of 5 shots in this sequence have Kitsu previews.

Facts

Shots

  • sc01_0010: Establishing the daunting scale of the human world, a static wide view surveys a 1920s garage filled with towering classic cars and…
  • sc01_0020: Panning across the gleaming chrome and dark green chassis of a classic automobile, the lens zeroes in on a ‘ROUTE 66’ badge, emphasizing…
  • sc01_0030: Grounded at the level of the asphalt, the camera holds on the tiny snail as a massive, sputtering Model T roars past in a blur, the sheer…
  • sc01_0040: Looking up from snail’s vulnerable vantage point, the towering tires of the Model T speed by, framing her tiny form against a backdrop…
  • sc01_0050: Pushing in slowly on the snail’s determined face, the blurred colors of an American flag and a Route 66 sign in the background isolate her…

sc01_0010

Establishing the daunting scale of the human world, a static wide view surveys a 1920s garage filled with towering classic cars and bustling people, culminating in a route-66 emblem that marks the beginning of the monumental path.

Preview

Latest Kitsu preview (revision 3, source: sc01_0010).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc01
  • status: running
  • canceled: no
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-01-B01
  • beat_summary: The snail looks up at the Route 66 sign and the endless road stretching to the horizon.
  • beat_shift: curiosity -> daunting realization
  • tempo: brisk
  • transition: Fade In

Camera

  • shot_size: Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 94
  • nb_frames: 94

Archive source

  • archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre#0031
  • archive_tc_in: 00:01:40:23
  • archive_tc_out: 100.92

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, CHICAGO - DAWN
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 3.75
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: era-locked-look. mood: Nostalgic, Quiet, Determined. time of day: dawn.
  • camera: Wide Shot; Eye Level; Static Shot; Rule of Thirds; Group Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip F719859_Kilimangiaro_Route_66_La_strada_madre to the cut. Regrade to match the 1920s Silent Film Era look, nostalgic and quiet at dawn. Retime to 3.75 seconds. Do not generate any new content.
  • reference_summary: A sequence of three shots showing the interior of a vintage car garage or museum, with several classic cars on display. The scene develops with slight angle changes and the entry of people interacting with the cars, culminating with the appearance of on-screen text recalling the Route 66 theme.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-01-B01
  • beat_summary: The snail observes the Route 66 sign and the vast road stretching to the horizon.
  • beat_shift: Quiet observation -> Dawning realization of scale
  • kind: archive
  • tempo: measured
  • duration_s: 5.0
  • editorial_note: Opens the sequence from black to establish the dawn setting and snail’s perspective.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0100
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 469.68
  • archive_out: 472.24
  • archive_tc_in: 00:07:49:17
  • archive_tc_out: 00:07:52:06
  • origin_code: arc
  • notes: Strong archive match (score 0.5364); relations: conceptual, lexical, narrative
  • description: A stylized map of the United States fills the frame from above, its deep focus capturing a red line steadily advancing from Chicago toward Los Angeles to establish the immense scale of the impending journey.
  • archive_summary_en: A sequence of three images showing a stylized map of Route 66 in the United States, with a red segment progressively moving along the route from Chicago to Los Angeles, indicating progress along the road. The logos and state names remain fixed, while only the position of the red segment changes, …

Notes

  • Part of sequence sc01 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: f9a9b0b6-0749-42bf-bfba-cf804a1ab46c
  • Kitsu entity id: d733e1e9-b5d9-4a03-ba99-a47eb463e7da (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc01_0020

Panning across the gleaming chrome and dark green chassis of a classic automobile, the lens zeroes in on a ‘ROUTE 66’ badge, emphasizing the intimidating mechanical giants that dominate this new environment.

Preview

Latest Kitsu preview (revision 1, source: sc01_0020).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc01
  • status: running
  • canceled: no
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-01-B01
  • beat_summary: The snail looks up at the route-66 sign and the endless road stretching to the horizon.
  • beat_shift: curiosity -> daunting realization
  • tempo: brisk
  • transition: Eyeline Match

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Eye Level
  • camera_movement: Pan
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 94
  • nb_frames: 94

Archive source

  • archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre#0115
  • archive_tc_in: 00:08:09:08
  • archive_tc_out: 489.32

Continuity

  • big framing jump (Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, CHICAGO - DAWN
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 3.75
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: era-locked-look. mood: Nostalgic, Quiet, Determined. time of day: dawn.
  • camera: Close-Up Shot; Eye Level; Pan; Leading Lines; Single Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip F719859_Kilimangiaro_Route_66_La_strada_madre to the cut, focusing on the pan across the dark green automobile and ‘ROUTE 66’ badge. Regrade to a 1920s Silent Film Era aesthetic, dawn lighting. Retime to 3.75 seconds. No generation.
  • reference_summary: A sequence of shots progressively showing the rear of a dark green automobile, with particular emphasis on the ‘ROUTE 66’ badge, the chrome strip and design details, including a taillight and a yellow sticker. The camera moves laterally, revealing more elements of the vehicle in each scene.
  • brief notes: big framing jump (Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-01-B01
  • beat_summary: The snail observes the Route 66 sign and the vast road stretching to the horizon.
  • beat_shift: Quiet observation -> Dawning realization of scale
  • kind: archive
  • tempo: measured
  • duration_s: 5.0
  • editorial_note: Cuts along her gaze to reveal the vastness of the road she is observing.
  • archive_segment_id: F719859_Kilimangiaro_Route_66_La_strada_madre#0016
  • source_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • archive_in: 44.32
  • archive_out: 46.96
  • archive_tc_in: 00:00:44:08
  • archive_tc_out: 00:00:46:24
  • origin_code: arc
  • notes: Strong archive match (score 0.5364); relations: conceptual, lexical, narrative | [4.6 re-select] plate re-selected -> F719859_Kilimangiaro_Route_66_La_strada_madre#0016 (score 0.52)
  • description: A static, head-on view of a black decorative plaque detailing the ‘America’s Main Street’ route from Chicago to Los Angeles, grounding the viewer in the historical reality of the road.
  • archive_summary_en: A sequence of three shots showing a yellow pickup truck with ‘Route US 66’ decorations moving on a road in a desert environment. The camera follows the vehicle from different angles: front, side, and rear, highlighting the driver and the landscape context.
  • revised: True
  • reselection: Roadbook#0054 -> F719859_Kilimangiaro_Route_66_La_strada_madre#0016

OMC sidecar (001__osc_sc01_0020_comp_gen_a002_v001_sf_rt66-map.1001.omc.json)

  • asset: osc_sc01_0020_comp_gen_a002_v001_sf_rt66-map.1001.png
  • deliverable stem: osc_sc01_0020_comp_cmp_a003_v001_sf_rt66-map.mov
  • era: Silent Film 1920s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1772080061
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T09:10:39.737Z
  • generation_seconds: 24
  • task_id: 019f12a3-b86c-7f60-8d98-b49c3ee75580
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc01 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 68d543be-ecd7-4f2c-ab10-67d8494ab517
  • Kitsu entity id: da2db778-33a3-4ca7-9a81-72a847058326 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc01_0030

Grounded at the level of the asphalt, the camera holds on the tiny snail as a massive, sputtering Model T roars past in a blur, the sheer force of its passing emphasizing her stark isolation against the sprawling, map-like expanse of the road.

Preview

Latest Kitsu preview (revision 1, source: sc01_0030).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc01
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-01-B02
  • beat_summary: A sputtering Model T Ford speeds past, completely ignoring snail’s existence.
  • beat_shift: vulnerability -> stark isolation
  • tempo: slow
  • transition: Smash Cut

Camera

  • shot_size: Macro Shot
  • camera_angle: Ground Level
  • camera_movement: Static Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 188
  • nb_frames: 188

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, CHICAGO - DAWN
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 7.5
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: era-locked-look. mood: Nostalgic, Quiet, Determined. time of day: dawn.
  • camera: Macro Shot; Ground Level; Static Shot; Rule of Thirds; Single Shot; Shallow Depth of Field; Wide Angle Lens; Slow Motion
  • prompt: A tiny, determined snail crawls on the rough asphalt at ground level. In the blurred background, a massive 1920s Model T car roars past in slow motion. The setting is rooted in the graphic, map-like textures of vintage route-66 Americana, with dawn lighting casting long shadows. Macro Shot, Ground Level, Static, Shallow Depth of Field, Wide Angle Lens, 1920s Silent Film Era aesthetic.
  • negative_prompt: modern vehicles, modern roads, bright midday sun, deep depth of field, fast motion, human faces, stylized map graphics on screen, text, water
  • reference_summary: A sequence of three images showing a stylized map of Route 66 in the United States, with a red segment progressively moving along the route from Chicago to Los Angeles, indicating progress along the road. The logos and state names remain fixed, while only the position of the red segment changes, …

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-01-B02
  • beat_summary: A smoking Model T Ford speeds past, completely oblivious to her presence.
  • beat_shift: Vulnerability -> Stark isolation
  • kind: synthetic
  • tempo: slow
  • duration_s: 10.0
  • editorial_note: Jarring contrast to her stillness as the speeding Model T violently interrupts the quiet.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Positioned at the gritty asphalt level inspired by the historic route’s origins, the camera violently whips to follow a roaring Model T Ford as it blasts past the tiny snail, emphasizing her extreme vulnerability against the massive, indifferent machine.

OMC sidecar (001__osc_sc01_0030_comp_gen_a002_v001_sf_model-t-ford.1001.omc.json)

  • asset: osc_sc01_0030_comp_gen_a002_v001_sf_model-t-ford.1001.png
  • deliverable stem: osc_sc01_0030_comp_cmp_a003_v001_sf_model-t-ford.mov
  • era: Silent Film 1920s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 128295265
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T09:08:46.679Z
  • generation_seconds: 13
  • task_id: 019f12a2-f022-7620-b30d-3b0f72b4d882
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc01 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: d49ac584-6bf2-49fc-a4d2-aca3bb96c826
  • Kitsu entity id: 15409554-0cf3-4394-8da2-376ebba5cff0 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc01_0040

Looking up from snail’s vulnerable vantage point, the towering tires of the Model T speed by, framing her tiny form against a backdrop reminiscent of vintage Americana road signs, highlighting her insignificance in the face of roaring progress.

Preview

Latest Kitsu preview (revision 1, source: sc01_0040).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc01
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-01-B02
  • beat_summary: A sputtering Model T Ford speeds past, completely ignoring the snail’s existence.
  • beat_shift: vulnerability -> stark isolation
  • tempo: slow
  • transition: Cut on Action

Camera

  • shot_size: Extreme Close-Up Shot
  • camera_angle: Worm’s Eye View
  • camera_movement: Tilt
  • lens: Ultra Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 188
  • nb_frames: 188

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, CHICAGO - DAWN
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 7.5
  • reference_archive_id: Roadbook
  • style: Era/look: era-locked-look. mood: Nostalgic, Quiet, Determined. time of day: dawn.
  • camera: Extreme Close-Up Shot; Worm’s Eye View; Tilt; Dynamic Symmetry; Single Shot; Rack Focus; Ultra Wide Angle Lens; Frame Rate
  • prompt: Extreme close-up from a worm’s eye view of a tiny snail looking up. Towering 1920s car tires speed by, framing the snail. The background textures evoke vintage black decorative plaques and classic Americana road signs. The camera tilts up with a rack focus from the snail to the roaring tires. Ultra Wide Angle Lens, dawn lighting, 1920s Silent Film Era look.
  • negative_prompt: modern tires, modern cars, daylight, shallow depth of field on tires, text, actual map on screen, plaque on screen, human figures
  • reference_summary: A static sequence showing a black decorative plaque with a map of the United States and route-66 road signs for different states, with the text ‘America’s Main Street’ in the center. The cities of Chicago and Los Angeles are indicated. In the lower right there is a round object with the word …

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-01-B03
  • beat_summary: The snail makes a quiet verbal commitment to herself and begins her crawl.
  • beat_shift: Hesitation -> Defiant resolve
  • kind: synthetic
  • tempo: slow
  • duration_s: 10.0
  • editorial_note: Punches in to a macro scale to capture her intimate, quiet commitment before moving.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Pushing in intimately on the snail’s face against the blurred backdrop of a vintage roadbook plaque, the shallow focus isolates her defiant expression as she physically leans forward to begin her monumental trek.

OMC sidecar (001__osc_sc01_0040_comp_gen_a002_v001_sf_rt66-sign.1001.omc.json)

  • asset: osc_sc01_0040_comp_gen_a002_v001_sf_rt66-sign.1001.png
  • deliverable stem: osc_sc01_0040_comp_cmp_a003_v001_sf_rt66-sign.mov
  • era: Silent Film 1920s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1574748063
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T09:49:51.935Z
  • generation_seconds: 15
  • task_id: 019f12c8-2456-7110-a0d5-deb97246c50b
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc01 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: af3aba2a-145e-4364-8257-98f0f1e4436f
  • Kitsu entity id: 81f76498-555e-4609-81d4-c38079994b93 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc01_0050

Pushing in slowly on snail’s determined face, the blurred colors of an American flag and a route-66 sign in the background isolate her quiet resolve as she takes her very first, deliberate slide forward.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc01
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-01-B03
  • beat_summary: The snail makes a quiet, private commitment and begins to crawl.
  • beat_shift: hesitation -> quiet resolve
  • tempo: slow
  • transition: Cut

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Eye Level
  • camera_movement: Push In
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 188
  • nb_frames: 188

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, CHICAGO - DAWN
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 7.5
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: era-locked-look. mood: Nostalgic, Quiet, Determined. time of day: dawn.
  • camera: Close-Up Shot; Eye Level; Push In; Centered; Single Shot; Bokeh; Medium Lens; Frame Rate
  • prompt: Close-up shot pushing in slowly on a determined snail’s face at eye level. The background features beautiful bokeh with the blurred, muted colors of a vintage American flag and a Route 66 sign. The snail takes a deliberate slide forward. Centered composition, Medium Lens, dawn lighting, 1920s Silent Film Era aesthetic.
  • negative_prompt: sharp background, modern flag, clear text, wide angle, static camera, deep depth of field, human figures, modern road
  • reference_summary: A static sequence showing a ‘ROUTE US 66’ road sign against a blurred American flag background. There are no changes between the scenes, maintaining a fixed and repetitive composition with slight digital artifacts visible.

Notes

  • Part of sequence sc01 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu entity id: 798cb52b-1155-47c5-ab5d-42ba82fdf687 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc02 - 02 The Shoulder & the Diner

Sequence sc02 - The Shoulder & the Diner

Preview

First available Kitsu preview for this sequence (from shot sc02_0010). 7 of 7 shots in this sequence have Kitsu previews.

Facts

Shots

  • sc02_0010: First-frame thumbnail shows a stylized, slug-like snail character (white shell, orange body, blue eye) on a cracked asphalt roadway, facing…
  • sc02_0020: First-frame thumbnail shows the slug-like snail character (orange body, white shell, blue eye, buck-toothed mouth) on cracked asphalt road…
  • sc02_0030: First-frame thumbnail shows a small, abstract snail-like creature (orange body, white shell, single visible blue eye) in the lower right…
  • sc02_0040: First-frame thumbnail shows a low-angle, eye-level shot from behind snail character: foreground is dominated by the underside of a…
  • sc02_0050: First-frame thumbnail shows the snail character (small orange body, white shell, colorful beach-ball-like head/markings) on a cracked…
  • sc02_0060: First-frame thumbnail shows a wide shot of the Ariston Cafe (Budweiser neon sign, large ‘ARISTON’ brick signage, Mobilgas pumps, brick…
  • sc02_0070: First-frame thumbnail shows a high-angle, pulled-back view of the cracked asphalt road shoulder, with the truck/trailer underside and wheel…

sc02_0010

First-frame thumbnail shows a stylized, slug-like snail character (white shell, orange body, blue eye) on a cracked asphalt roadway, facing away from camera. A blurred vehicle occupies the right third of the frame (suggesting a passing car to be composited in shot), and a small bush sits on the curb at lower right. Daylight, mid-1950s roadside diner context.

Preview

Latest Kitsu preview (revision 2, source: osc_sc02_0010_comp_gen_a002_v002_nr_ariston_cafe.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc02
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-02-B01
  • beat_summary: The snail travels along the shoulder as confident mid-century traffic, including Cadillacs and a Harley, roars past.
  • beat_shift: isolation -> rhythmic persistence
  • tempo: measured
  • transition: Cross Dissolve

Camera

  • shot_size: Medium Close-Up Shot
  • camera_angle: Ground Level
  • camera_movement: Tracking Shot
  • lens: Anamorphic Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 125
  • nb_frames: 125

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, MIDWEST - DAY / EXT. GAS STATION - DUSK
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 5.0
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: Mid-Century (1940s-1950s). mood: Bustling, Confident, Rhythmic. time of day: day / ext. gas station - dusk.
  • camera: Medium Close-Up Shot; Ground Level; Tracking Shot; Leading Lines; Single Shot; Shallow Depth of Field; Anamorphic Medium Lens; Frame Rate
  • prompt: Medium close-up tracking shot at ground level moving smoothly alongside a snail on a gravel shoulder. In the shallow depth of field background, the gleaming chrome bumpers and vibrant fins of 1950s classic cars blur past. The textures and lighting reflect a bustling mid-century day transitioning to dusk. Anamorphic Medium Lens, leading lines.
  • negative_prompt: modern cars, static camera, deep depth of field, interior garage, museum setting, text, human figures, modern asphalt
  • reference_summary: A sequence of three shots showing the interior of a vintage car garage or museum, with several classic cars on display. The scene develops with slight angle changes and the entry of people interacting with the cars, culminating with the appearance of on-screen text recalling the route-66 theme.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-02-B01
  • beat_summary: The snail crawls along the shoulder as confident mid-century traffic roars past.
  • beat_shift: Isolation -> Immersion in the flow
  • kind: synthetic
  • tempo: measured
  • duration_s: 5.38
  • editorial_note: Blends the eras to signify the passage of time into the bustling 1940s.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Tracking smoothly alongside snail at ground level, the composition uses the endless edge of the highway as a leading line, framing her steady progress against the towering, blurred rush of mid-century chrome and steel.

OMC sidecar (001__osc_sc02_0010_comp_gen_a002_v001_nr_gas_station.1001.omc.json)

  • asset: osc_sc02_0010_comp_gen_a002_v001_nr_gas_station.1001.png
  • deliverable stem: osc_sc02_0010_comp_cmp_a003_v001_nr_gas_station.mov
  • era: Newsreel 1950s-60s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1121207376
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:03:43.511Z
  • generation_seconds: 12
  • task_id: 019f12d5-5a2e-73b3-be20-b5d23cbdc850
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc02 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 0e0b331a-da93-4658-b156-fc150dd60f66
  • Kitsu entity id: f6a76ff6-b485-4915-9f0a-a3db416c5605 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc02_0020

First-frame thumbnail shows the slug-like snail character (orange body, white shell, blue eye, buck-toothed mouth) on cracked asphalt road surface, with a red Mobilgas shield visible in the upper left and warm golden-hour lighting. Wide-angle composition emphasizing the textured road shoulder and the small scale of the character against the broad pavement.

Preview

Latest Kitsu preview (revision 2, source: osc_sc02_0020_comp_gen_a002_v002_nr_snail_roadside.1001.png).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc02
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-02-B01
  • beat_summary: The snail travels along the shoulder as confident mid-century traffic, including Cadillacs and a Harley, roars past.
  • beat_shift: isolation -> rhythmic persistence
  • tempo: measured
  • transition: Cut on Action

Camera

  • shot_size: Full Shot
  • camera_angle: Low Angle
  • camera_movement: Dolly Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 125
  • nb_frames: 125

Continuity

  • big framing jump (Medium Close-Up Shot -> Full Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, MIDWEST - DAY / EXT. GAS STATION - DUSK
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 5.0
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: Mid-Century (1940s-1950s). mood: Bustling, Confident, Rhythmic. time of day: day / ext. gas station - dusk.
  • camera: Full Shot; Low Angle; Dolly Shot; Rule of Thirds; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Full shot from a low angle dollying alongside a snail making steady progress across rocky, arid roadside terrain. In the deep depth of field background, a 1950s motorcycle speeds by against a vast, transitioning desert landscape. Wide Angle Lens, Rule of Thirds, mid-century 1950s aesthetic, dusk lighting.
  • negative_prompt: shallow depth of field, static camera, modern motorcycles, lush green forest, interior car view, window frame, text, human faces
  • reference_summary: The sequence shows a view from the front window of a moving car through a green natural environment, which then transforms into an arid and rocky landscape, with an increase in visual disturbances in the last scene.
  • brief notes: big framing jump (Medium Close-Up Shot -> Full Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-02-B01
  • beat_summary: The snail crawls along the shoulder as confident mid-century traffic roars past.
  • beat_shift: Isolation -> Immersion in the flow
  • kind: synthetic
  • tempo: measured
  • duration_s: 5.38
  • editorial_note: Maintains her continuous crawling motion as the camera shifts from tracking to a dolly move.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Dollying forward with snail as she navigates the dusty shoulder, the shallow depth of field keeps her in sharp focus while the bustling, sunlit Americana of a roadside store’s exterior blurs warmly in the background.

OMC sidecar (001__osc_sc02_0020_comp_gen_a002_v001_nr_roadside_signs.1001.omc.json)

  • asset: osc_sc02_0020_comp_gen_a002_v001_nr_roadside_signs.1001.png
  • deliverable stem: osc_sc02_0020_comp_cmp_a003_v001_nr_roadside_signs.mov
  • era: Newsreel 1950s-60s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1134065543
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:09:13.897Z
  • generation_seconds: 14
  • task_id: 019f12da-15cb-74c3-bc1b-f179345d1b4e
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc02 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: cd4c4248-9ccc-4a9f-871e-af309b0aba38
  • Kitsu entity id: 14d7943a-59bd-4146-9608-733c0e9e5786 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc02_0030

First-frame thumbnail shows a small, abstract snail-like creature (orange body, white shell, single visible blue eye) in the lower right foreground on a cracked asphalt road, with a blurred yellow vehicle/machinery occupying the left third of the frame and a hazy horizon at the top. Daylight, ground-level camera angle, 1950s roadside setting.

Preview

Latest Kitsu preview (revision 2, source: osc_sc02_0030_comp_gen_a002_v002_nr_snail_cars.1001.png).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc02
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-02-B01
  • beat_summary: The snail travels along the shoulder as confident mid-century traffic, including Cadillacs and a Harley, roars past.
  • beat_shift: isolation -> rhythmic persistence
  • tempo: measured
  • transition: Cut

Camera

  • shot_size: Medium Shot
  • camera_angle: Ground Level
  • camera_movement: Orbiting Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 125
  • nb_frames: 125

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, MIDWEST - DAY / EXT. GAS STATION - DUSK
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 5.0
  • reference_archive_id: T08340_222_TG2_SI_VIAGGIARE_2008__2009_Viaggio_lungo_la_Route_66
  • style: Era/look: Mid-Century (1940s-1950s). mood: Bustling, Confident, Rhythmic. time of day: day / ext. gas station - dusk.
  • camera: Medium Shot; Ground Level; Orbiting Shot; Dynamic Symmetry; Single Shot; Rack Focus; Medium Lens; Frame Rate
  • prompt: Medium shot at ground level orbiting around a snail navigating a dusty roadside. A rack focus shifts to a vintage 1950s motorcycle speeding by in the background, contrasting snail’s slow pace. The environment features coastal mid-century textures, hinting at a distant pier and sea. Dynamic symmetry, Medium Lens, dusk lighting.
  • negative_prompt: static camera, deep depth of field, modern motorcycles, modern clothing, clear text, Santa Monica sign on screen, modern vehicles
  • reference_summary: The sequence begins with a close-up of a Santa Monica pier sign, then moves to a view along the pier with people and a moving motorcycle, showing the port and the sea in the background. The transition is fluid, with a slight shift in perspective to the right in the third scene.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-02-B01
  • beat_summary: The snail crawls along the shoulder as confident mid-century traffic roars past.
  • beat_shift: Isolation -> Immersion in the flow
  • kind: synthetic
  • tempo: measured
  • duration_s: 5.38
  • editorial_note: Widens the frame to juxtapose her slow progress against the roaring mid-century traffic.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Panning slowly across the vast, arid desert landscape, the wide lens captures the snail’s tiny silhouette pressing forward into the expansive leading room, dwarfed by the historic trading post signs and the roaring traffic.

Notes

  • Part of sequence sc02 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: d50aa07a-4b52-4cbb-92ae-36abc105d191
  • Kitsu entity id: ae01190c-f6db-4c30-b08a-5ebba6e1673c (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc02_0040

First-frame thumbnail shows a low-angle, eye-level shot from behind snail character: foreground is dominated by the underside of a large vehicle (tractor-trailer or truck with visible wheel treads), silhouetted against a warm desert horizon, with a small scrub bush visible on the left. Golden-hour lighting, dust on the road shoulder, no diner visible yet.

Preview

Latest Kitsu preview (revision 2, source: osc_sc02_0040_comp_gen_a002_v002_nr_wagon_wheel.1001.png).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc02
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-02-B02
  • beat_summary: The snail slows down to listen to music drifting out from a roadside diner.
  • beat_shift: relentless motion -> fleeting connection
  • tempo: rapid
  • transition: Cross Dissolve

Camera

  • shot_size: Macro Shot
  • camera_angle: Eye Level
  • camera_movement: Push In
  • lens: Telephoto Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 62
  • nb_frames: 62

Continuity

  • big framing jump (Medium Shot -> Macro Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, MIDWEST - DAY / EXT. GAS STATION - DUSK
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.5
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: Mid-Century (1940s-1950s). mood: Bustling, Confident, Rhythmic. time of day: day / ext. gas station - dusk.
  • camera: Macro Shot; Eye Level; Push In; Centered; Single Shot; Bokeh; Telephoto Lens; Slow Motion
  • prompt: Macro shot pushing in gently on a snail at eye level as it halts its crawl, antennae perking up. The background is a heavy bokeh of warm, neon-lit glows from a distant 1950s roadside diner, with colors reminiscent of vintage Americana. Centered composition, Telephoto Lens, slow motion, mid-century dusk aesthetic.
  • negative_prompt: sharp background, deep depth of field, wide angle, fast motion, clear text, American flag on screen, modern neon, daylight
  • reference_summary: A static sequence showing a ‘ROUTE US 66’ road sign against a blurred American flag background. There are no changes between the scenes, maintaining a fixed and repetitive composition with slight digital artifacts visible.
  • brief notes: big framing jump (Medium Shot -> Macro Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-02-B02
  • beat_summary: She pauses by a diner, closing her eyes to listen to the drifting radio music.
  • beat_shift: Relentless motion -> Momentary peace
  • kind: archive
  • tempo: rapid
  • duration_s: 2.69
  • editorial_note: Drops into a rapid close-up to match the sudden pause in her movement by the diner.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0527
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 2986.04
  • archive_out: 2989.92
  • archive_tc_in: 00:49:46:01
  • archive_tc_out: 00:49:49:23
  • origin_code: arc
  • notes: Strong archive match (score 0.5174); relations: conceptual, lexical, narrative | [4.6 re-select] plate re-selected -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0527 (score 0.516)
  • description: A static, centered close-up of a classic ‘ROUTE US 66’ sign standing proudly against a softly blurred American flag, evoking the nostalgic, patriotic atmosphere of the era’s roadside culture.
  • archive_summary_en: The sequence shows a beige car and a silver caravan parked on arid ground, with a scratched wall in the background. The camera pans from left to right, progressively highlighting the caravan, which becomes the main object of the final scene.
  • revised: True
  • reselection: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0009 -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0527

OMC sidecar (001__osc_sc02_0040_comp_gen_a002_v001_nr_snail_listen.1001.omc.json)

  • asset: osc_sc02_0040_comp_gen_a002_v001_nr_snail_listen.1001.png
  • deliverable stem: osc_sc02_0040_comp_cmp_a003_v001_nr_snail_listen.mov
  • era: Newsreel 1950s-60s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1401098916
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:19:45.314Z
  • generation_seconds: 13
  • task_id: 019f12e3-ddc6-7791-b931-f5786ca1eeee
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc02 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 2b8116c8-26ed-4039-a5a5-fc936904ad5b
  • Kitsu entity id: a1d32759-434d-4e8c-9bcd-433f0da42e7b (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc02_0050

First-frame thumbnail shows snail character (small orange body, white shell, colorful beach-ball-like head/markings) on a cracked asphalt road, with a Mobilgas-branded service station in the background: two red Mobilgas pumps, a brick storefront with ‘Mobilgas’ shields, a mailbox, and an orange traffic cone. Warm daylight, mid-century roadside Americana, eye-level static shot.

Preview

Latest Kitsu preview (revision 2, source: osc_sc02_0050_comp_gen_a002_v002_nr_truck_wheel.1001.png).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc02
  • status: running
  • canceled: no
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-02-B02
  • beat_summary: The snail slows down to listen to music drifting out from a roadside diner.
  • beat_shift: relentless motion -> fleeting connection
  • tempo: rapid
  • transition: Eyeline Match

Camera

  • shot_size: Medium Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 62
  • nb_frames: 62

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0076
  • archive_tc_in: 00:06:18:04
  • archive_tc_out: 378.16

Continuity

  • big framing jump (Macro Shot -> Medium Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, MIDWEST - DAY / EXT. GAS STATION - DUSK
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 2.5
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: Mid-Century (1940s-1950s). mood: Bustling, Confident, Rhythmic. time of day: day / ext. gas station - dusk.
  • camera: Medium Shot; Eye Level; Static Shot; Rule of Thirds; Two-Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a warm Mid-Century (1940s-1950s) dusk look, focusing on the medium shot of the clerk and customer beneath the route-66 hat. Retime to 2.5 seconds. No generation.
  • reference_summary: A sequence of interactions in a store, with John Smith behind the counter assisting Lisa Brown during a purchase, while Mary Jones is initially present but no longer in subsequent scenes. The environment is a store with displayed items, natural lighting and a ‘Route 66’ hat hanging from the ceiling.
  • brief notes: big framing jump (Macro Shot -> Medium Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-02-B02
  • beat_summary: She pauses by a diner, closing her eyes to listen to the drifting radio music.
  • beat_shift: Relentless motion -> Momentary peace
  • kind: archive
  • tempo: rapid
  • duration_s: 2.69
  • editorial_note: Pulls back slightly to contextualize her resting state as she listens to the music.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0076
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 378.16
  • archive_out: 381.16
  • archive_tc_in: 00:06:18:04
  • archive_tc_out: 00:06:21:04
  • origin_code: arc
  • notes: Strong archive match (score 0.5174); relations: conceptual, lexical, narrative
  • description: A warm, natural-light medium shot captures the everyday Americana of a roadside store, framing a shopkeeper and a customer interacting across the counter beneath a hanging Route 66 hat.
  • archive_summary_en: A sequence of interactions in a store, with John Smith behind the counter assisting Lisa Brown during a purchase, while Mary Jones is present initially but no longer in the subsequent scenes. The environment is a store with displayed items, natural lighting, and a ‘Route 66’ hat hanging from the …

OMC sidecar (001__osc_sc02_0050_comp_gen_a002_v001_nr_trucker_break.1001.omc.json)

  • asset: osc_sc02_0050_comp_gen_a002_v001_nr_trucker_break.1001.png
  • deliverable stem: osc_sc02_0050_comp_cmp_a003_v001_nr_trucker_break.mov
  • era: Newsreel 1950s-60s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1219789483
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:26:33.372Z
  • generation_seconds: 11
  • task_id: 019f12e9-9300-7ec3-b807-faeb58534e28
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc02 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 0951b7f1-2bea-4eff-8905-9b84635627f8
  • Kitsu entity id: 7b1ae0fa-7fe3-4c20-9d62-d8ae87b07eec (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc02_0060

First-frame thumbnail shows a wide shot of the Ariston Cafe (Budweiser neon sign, large ‘ARISTON’ brick signage, Mobilgas pumps, brick arches, globe lights on roof) on route-66. The snail character with white shell and orange body stands in the center foreground beneath the ‘ARISTON’ sign, between the two red Mobilgas pumps. Warm golden-hour lighting, mid-century Americana facade.

Preview

Latest Kitsu preview (revision 2, source: osc_sc02_0060_comp_gen_a002_v002_nr_snail_mobilgas.1001.png).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc02
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-02-B03
  • beat_summary: A trucker steps out of the diner, breaking the moment and prompting snail to move on.
  • beat_shift: distraction -> renewed focus
  • tempo: measured
  • transition: Cut on Action

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Ground Level
  • camera_movement: Whip Pan
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 125
  • nb_frames: 125

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, MIDWEST - DAY / EXT. GAS STATION - DUSK
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 5.0
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: Mid-Century (1940s-1950s). mood: Bustling, Confident, Rhythmic. time of day: day / ext. gas station - dusk.
  • camera: Close-Up Shot; Ground Level; Whip Pan; Leading Room; Single Shot; Shallow Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Close-up shot at ground level featuring a sudden whip pan from a dusty diner entrance to a snail on arid dirt. A heavy 1950s trucker’s boot thuds onto the gravel right next to the snail. The background is a blurred, vast desert landscape with mid-century textures. Shallow depth of field, Wide Angle Lens, leading room, dusk lighting.
  • negative_prompt: static camera, deep depth of field, modern shoes, modern diner, clear text, historical sign on screen, modern vehicles
  • reference_summary: The sequence begins with a vast desert landscape with a bridge in the distance, moves to a historical sign indicating one of the oldest trading posts on Route 66, and ends with a distorted and blurred vision of the desert, suggesting a journey or transition through an arid and isolated environment.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-02-B02
  • beat_summary: She pauses by a diner, closing her eyes to listen to the drifting radio music.
  • beat_shift: Relentless motion -> Momentary peace
  • kind: archive
  • tempo: rapid
  • duration_s: 2.69
  • editorial_note: Expands to an extreme wide to emphasize the drifting radio tunes in the vast landscape.
  • archive_segment_id: F719859_Kilimangiaro_Route_66_La_strada_madre#0031
  • source_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • archive_in: 100.92
  • archive_out: 105.96
  • archive_tc_in: 00:01:40:23
  • archive_tc_out: 00:01:45:24
  • origin_code: arc
  • notes: Strong archive match (score 0.5174); relations: conceptual, lexical, narrative | [4.6 re-select] plate re-selected -> F719859_Kilimangiaro_Route_66_La_strada_madre#0031 (score 0.516)
  • description: A sweeping pan across the sun-baked desert reveals a distant bridge and a historic trading post sign, capturing the vast, isolated beauty of the mid-century highway landscape.
  • archive_summary_en: A sequence of three shots showing the interior of a garage or vintage car museum, with several classic cars on display. The scene develops with slight angle changes and the entry of people interacting with the cars, culminating with the appearance of text on the screen recalling the Route 66 theme.
  • revised: True
  • reselection: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0281 -> F719859_Kilimangiaro_Route_66_La_strada_madre#0031

OMC sidecar (001__osc_sc02_0060_comp_gen_a002_v001_nr_resume_walk.1001.omc.json)

  • asset: osc_sc02_0060_comp_gen_a002_v001_nr_resume_walk.1001.png
  • deliverable stem: osc_sc02_0060_comp_cmp_a003_v001_nr_resume_walk.mov
  • era: Newsreel 1950s-60s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1746443256
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:34:28.024Z
  • generation_seconds: 15
  • task_id: 019f12f1-6558-71c3-b35a-8cac8a09a217
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc02 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 5baddf25-1303-4cc4-8e1b-d14fd49c268d
  • Kitsu entity id: bfc53a4a-67ab-4c87-abb7-a8c17bdbb968 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc02_0070

First-frame thumbnail shows a high-angle, pulled-back view of the cracked asphalt road shoulder, with the truck/trailer underside and wheel still prominent in the upper right (shadow falling across the road), a small green plant on the left curb, and a single small white object (possibly snail or a marker) in the lower right corner. Dusty roadside, warm lighting, no signage visible.

Preview

Latest Kitsu preview (revision 1, source: osc_sc02_0070_comp_gen_a002_v001_nr_ariston_cafe.1001.png).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc02
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-02-B03
  • beat_summary: A trucker steps out of the diner, breaking the moment and prompting the snail to move on.
  • beat_shift: distraction -> renewed focus
  • tempo: measured
  • transition: Cut on Action

Camera

  • shot_size: Full Shot
  • camera_angle: High Angle
  • camera_movement: Pull Out
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 125
  • nb_frames: 125

Continuity

  • big framing jump (Close-Up Shot -> Full Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, MIDWEST - DAY / EXT. GAS STATION - DUSK
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 5.0
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: Mid-Century (1940s-1950s). mood: Bustling, Confident, Rhythmic. time of day: day / ext. gas station - dusk.
  • camera: Full Shot; High Angle; Pull Out; Rule of Thirds; Single Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Full shot from a high angle pulling out slightly as a trucker’s shadow falls over a snail. The snail resumes its rhythmic slide past the weathered wooden post of a vintage road sign. The deep depth of field reveals a rural 1950s highway landscape with a silver caravan in the distance. Rule of Thirds, Medium Lens, mid-century dusk.
  • negative_prompt: shallow depth of field, low angle, push in, modern vehicles, clear text ‘Old 66 Life’, interior car view, modern road
  • reference_summary: The sequence begins with a shot from a car window following a silver caravan on a rural road, then moves to a road sign with the text ‘Old 66 Life’, maintained for subsequent scenes with slight background variations, suggesting a journey along the historic route-66.
  • brief notes: big framing jump (Close-Up Shot -> Full Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-02-B03
  • beat_summary: A trucker exits the diner, breaking the spell and prompting her to move on.
  • beat_shift: Distraction -> Renewed focus
  • kind: synthetic
  • tempo: measured
  • duration_s: 5.38
  • editorial_note: Cuts on the trucker’s movement exiting the diner to abruptly break the tranquil spell.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Looking up past the snail’s shell, the focus racks sharply from her resting form to the heavy boots of a trucker stepping out of a diner, the sudden vertical motion breaking her reverie and forcing her back into reality.

Notes

  • Part of sequence sc02 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: ba37eb0d-0556-4761-b005-90d92b66b047
  • Kitsu entity id: 7747062f-924d-47ef-8734-885fde6c0cda (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc03 - 03 The Rain & the TV

Sequence sc03 - The Rain & the TV

Preview

First available Kitsu preview for this sequence (from shot sc03_0010). 8 of 10 shots in this sequence have Kitsu previews.

Facts

Shots

  • sc03_0010: As a pickup truck speeds through a darkening, rain-swept landscape, the camera tilts down to reveal snail desperately inching her way…
  • sc03_0020: Framed against the chrome engine details of a classic car parked under the station’s canopy, the snail huddles near the warm, dry tire, her…
  • sc03_0030: A discarded newspaper announcing the lunar mission lies on the wet concrete; the camera pushes in on the snail as she takes shelter beneath…
  • sc03_0040: Superimposed over a stylized map of route-66 painted on the station wall, the snail rests motionless, her steady progress temporarily…
  • sc03_0050: The camera slowly tracks around the snail as she gazes up at a glowing circular window, the reflection of a flickering television screen…
  • sc03_0060 (canceled): A black-and-white broadcast on the television screen shows a bustling city crowd, the camera pushing in on the human faces to emphasize the…
  • sc03_0070 (canceled): The television broadcast cuts to a vast, distorted desert landscape along Route 66, juxtaposing the earthly isolation of the historic…
  • sc03_0080 (canceled): A sweeping pan across a rocky, rain-washed landscape reveals a sudden burst of lush vegetation and a shimmering pool of water, as vibrant…
  • sc03_0090 (canceled): Pushing in from a wide view of a vibrant, neon-lit Route 66 diner interior to a close-up of a hanging toy car, the rich, saturated colors…
  • sc03_0100 (canceled): A glowing neon sign on a city building bursts into brilliant color as a vehicle passes, the radiant light cutting through the lingering…

sc03_0010

As a pickup truck speeds through a darkening, rain-swept landscape, the camera tilts down to reveal snail desperately inching her way toward the dry concrete beneath a gas station overhang, seeking refuge from the sudden downpour.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0010_comp_gen_a002_v001_kc_gas_station_rainbow.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: no
  • kind: composite
  • reference_archive: cmp
  • beat: SCENE-03-B01
  • beat_summary: Heavy rain forces the snail to seek shelter under a gas station overhang.
  • beat_shift: steady progress -> forced pause
  • tempo: brisk
  • transition: Cross Dissolve

Camera

  • shot_size: Medium Wide Shot
  • camera_angle: High Angle
  • camera_movement: Tilt
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 103
  • nb_frames: 103

Continuity

  • composite 1/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: composite
  • method: Composite - generate the authored subject and fuse it onto the cited real archive plate.
  • duration_s: 4.12
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Medium Wide Shot; High Angle; Tilt; Leading Lines; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Composite a tiny, desperate snail inching toward dry concrete beneath a gas station overhang onto the provided archive plate of a pickup truck speeding through a darkening, rain-swept landscape. The camera tilts down to reveal the snail. Match the 1969 late afternoon/evening atmospheric, awe-struck mood. Medium Wide Shot, High Angle, Deep Depth of Field, Wide Angle Lens.
  • negative_prompt: cartoon snail, mismatched lighting, dry environment, modern vehicles, shallow depth of field, static camera, wind turbines on screen
  • reference_summary: The sequence begins with a panoramic view of wind turbines in an arid landscape, then moves to a pickup truck traveling through a wooded area, ending with a progressive distortion and blurring of the image, until a complete loss of definition.
  • brief notes: composite 1/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B01
  • beat_summary: The snail takes shelter under a gas station overhang during a sudden, heavy downpour.
  • beat_shift: Exposure -> Cautious safety
  • kind: composite
  • tempo: rapid
  • duration_s: 2.32
  • editorial_note: Shifts the era to 1969 with a smooth transition into the sudden, heavy downpour.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0020
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 106.0
  • archive_out: 107.6
  • archive_tc_in: 00:01:46:00
  • archive_tc_out: 00:01:47:15
  • origin_code: cmp
  • notes: Augment archive F157301_DREAMS_ROADS___STREET_OF_AMERICA with AI/VFX (score 0.519) | [4.6 re-select] plate re-selected -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0020 (score 0.5)
  • description: Composited onto the edge of a weathered ‘Old 66 Life’ road sign, the snail clings tightly to the metal surface as the camera tracks past, finding a momentary anchor amidst the passing blur of a silver caravan on the rural road.
  • archive_summary_en: A static sequence showing a white truck with a cylindrical tank parked along a road, with a background of bare trees and a light-colored building under an overcast sky. There are no changes between the scenes, nor any detectable movements or actions.
  • revised: True
  • reselection: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0326 -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0020

OMC sidecar (001__osc_sc03_0010_comp_gen_a002_v001_kc_gas_station_rainbow.1001.omc.json)

  • asset: osc_sc03_0010_comp_gen_a002_v001_kc_gas_station_rainbow.1001.png
  • deliverable stem: osc_sc03_0010_comp_cmp_a003_v001_kc_gas_station_rainbow.mov
  • era: era-locked-look 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1859936254
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:46:17.987Z
  • generation_seconds: 14
  • task_id: 019f12fc-3a83-72f0-ab0b-9ebe8761454d
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Composite of archive + generated layers.
  • Kitsu preview file id: c6980234-0e80-4d30-9d8f-52a27a5175d9
  • Kitsu entity id: 8d7b55ec-d367-46d6-a3bc-e5cc5a589baa (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:51.

sc03_0020

Framed against the chrome engine details of a classic car parked under the station’s canopy, snail huddles near the warm, dry tire, her vulnerability highlighted as two men sit safely inside the vehicle, oblivious to the storm outside.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0020_comp_gen_a002_v001_kc_puddle_rainbow.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: no
  • kind: composite
  • reference_archive: cmp
  • beat: SCENE-03-B01
  • beat_summary: Heavy rain forces the snail to seek shelter under a gas station overhang.
  • beat_shift: steady progress -> forced pause
  • tempo: brisk
  • transition: Match Cut

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Ground Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 103
  • nb_frames: 103

Continuity

  • composite 2/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make; big framing jump (Medium Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: composite
  • method: Composite - generate the authored subject and fuse it onto the cited real archive plate.
  • duration_s: 4.12
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Close-Up Shot; Ground Level; Static Shot; Rule of Thirds; Single Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Composite a vulnerable snail huddled near a warm, dry car tire onto the provided archive plate of a classic car with two men sitting safely inside. The snail is framed against the chrome engine details under a gas station canopy during a rainstorm. Match the era-locked-look evening look. Close-Up Shot, Ground Level, Static, Shallow Depth of Field, Medium Lens.
  • negative_prompt: cartoon snail, mismatched lighting, dry environment, modern car, deep depth of field, moving camera, bright daylight
  • reference_summary: The sequence begins with a detail of a classic car engine, then moves inside the cabin where two men are seated, one at the wheel and the other next to him, in an environment that looks like a classic car garage or showroom. The scene remains inside the car with a slight variation in angle or …
  • brief notes: composite 2/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make; big framing jump (Medium Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B01
  • beat_summary: The snail takes shelter under a gas station overhang during a sudden, heavy downpour.
  • beat_shift: Exposure -> Cautious safety
  • kind: composite
  • tempo: rapid
  • duration_s: 2.32
  • editorial_note: Moves to a medium close-up to focus on her taking shelter from the aggressive rain.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0545
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 3185.16
  • archive_out: 3194.88
  • archive_tc_in: 00:53:05:04
  • archive_tc_out: 00:53:14:22
  • origin_code: cmp
  • notes: Augment archive F157301_DREAMS_ROADS___STREET_OF_AMERICA with AI/VFX (score 0.519) | [4.6 re-select] plate re-selected -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0545 (score 0.519)
  • description: The snail is fused onto the base of an ‘Old. 66 Cafe’ sign, her small form holding perfectly still against the metal pole as digital disturbances flicker across the official route markers above her.
  • archive_summary_en: The sequence shows a progression from an Airstream caravan in an open environment, with people entering and exiting, to a more remote and wild scene, where human figures are present in a blurred natural context, with obvious visual defects in the third scene.
  • revised: True
  • reselection: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0327 -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0545

OMC sidecar (001__osc_sc03_0020_comp_gen_a002_v001_kc_puddle_rainbow.1001.omc.json)

  • asset: osc_sc03_0020_comp_gen_a002_v001_kc_puddle_rainbow.1001.png
  • deliverable stem: osc_sc03_0020_comp_cmp_a003_v001_kc_puddle_rainbow.mov
  • era: Kodachrome 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 487585715
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:54:41.255Z
  • generation_seconds: 18
  • task_id: 019f1302-b2b7-7402-8089-b2747d3acb6b
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Composite of archive + generated layers.
  • Kitsu preview file id: c462bead-f08a-484b-9218-6fbc74bb4d8f
  • Kitsu entity id: 1bbbadc8-11da-48a3-a24b-91ac28b2b459 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0030

A discarded newspaper announcing the lunar mission lies on the wet concrete; the camera pushes in on snail as she takes shelter beneath its curling, rain-splattered edge, finding a forced pause amidst the monumental news of the era.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0030_comp_gen_a002_v001_kc_color_bleed.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: no
  • kind: composite
  • reference_archive: cmp
  • beat: SCENE-03-B01
  • beat_summary: Heavy rain forces the snail to seek shelter under a gas station overhang.
  • beat_shift: steady progress -> forced pause
  • tempo: brisk
  • transition: Cross Dissolve

Camera

  • shot_size: Macro Shot
  • camera_angle: Low Angle
  • camera_movement: Push In
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 103
  • nb_frames: 103

Continuity

  • composite 3/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: composite
  • method: Composite - generate the authored subject and fuse it onto the cited real archive plate.
  • duration_s: 4.12
  • reference_archive_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Macro Shot; Low Angle; Push In; Leading Lines; Single Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Composite a snail taking shelter beneath the curling, rain-splattered edge of a discarded 1969 newspaper onto the provided archive plate. The camera pushes in from a low angle on the snail. The newspaper features a lunar mission announcement. Match the atmospheric, awe-struck evening mood. Macro Shot, Shallow Depth of Field, Medium Lens.
  • negative_prompt: cartoon snail, mismatched lighting, dry newspaper, deep depth of field, static camera, modern text, rocket launch on screen
  • reference_summary: The sequence begins with a newspaper announcing a lunar event, continues with the launch of a rocket and ends with an astronaut on the lunar surface, creating a visual narrative of space travel from the moment of the announcement to the landing.
  • brief notes: composite 3/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B01
  • beat_summary: The snail takes shelter under a gas station overhang during a sudden, heavy downpour.
  • beat_shift: Exposure -> Cautious safety
  • kind: composite
  • tempo: rapid
  • duration_s: 2.32
  • editorial_note: Pushes in to build a sense of intimacy and isolation during the storm.
  • archive_segment_id: C5521_STORIA_DELL__INFORMAZIONE#0057
  • source_archive_id: C5521_STORIA_DELL__INFORMAZIONE
  • archive_in: 466.92
  • archive_out: 468.52
  • archive_tc_in: 00:07:46:23
  • archive_tc_out: 00:07:48:13
  • origin_code: cmp
  • notes: Augment archive F157301_DREAMS_ROADS___STREET_OF_AMERICA with AI/VFX (score 0.519) | [4.6 re-select] plate re-selected -> C5521_STORIA_DELL__INFORMAZIONE#0057 (score 0.5)
  • description: Pushing in low along the center line of a deserted, arid highway, the snail is composited directly onto the cracked asphalt, her slow progress contrasting with the graphic map overlays and the endless road stretching to the horizon.
  • archive_summary_en: The sequence begins with a rural view of a dirt road among vegetation and houses, moves to a blurred and dynamic scene with a central tree, and concludes with a cross-shaped pole against a cloudy sky background. The transition between scenes suggests rapid movement or a context shift, from a …
  • revised: True
  • reselection: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0406 -> C5521_STORIA_DELL__INFORMAZIONE#0057

OMC sidecar (001__osc_sc03_0030_comp_gen_a002_v001_kc_color_bleed.1001.omc.json)

  • asset: osc_sc03_0030_comp_gen_a002_v001_kc_color_bleed.1001.png
  • deliverable stem: osc_sc03_0030_comp_cmp_a003_v001_kc_color_bleed.mov
  • era: era-locked-look 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 101243556
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T10:57:53.268Z
  • generation_seconds: 12
  • task_id: 019f1306-acfd-7ee3-ab74-e246e760dd59
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Composite of archive + generated layers.
  • Kitsu preview file id: c1041b0b-4bb6-4e53-8806-089e2db02933
  • Kitsu entity id: 4bf86c0e-1e44-4833-a86c-4fb133a7ee34 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0040

Superimposed over a stylized map of route-66 painted on the station wall, snail rests motionless, her steady progress temporarily halted by the rain, contrasting her tiny physical presence with the vast red line of the journey ahead.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0040_comp_gen_a002_v001_kc_snail_puddle.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: no
  • kind: composite
  • reference_archive: cmp
  • beat: SCENE-03-B01
  • beat_summary: Heavy rain forces the snail to seek shelter under a gas station overhang.
  • beat_shift: steady progress -> forced pause
  • tempo: brisk
  • transition: Cut on Action

Camera

  • shot_size: Full Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 103
  • nb_frames: 103

Continuity

  • composite 4/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make; big framing jump (Macro Shot -> Full Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: composite
  • method: Composite - generate the authored subject and fuse it onto the cited real archive plate.
  • duration_s: 4.12
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Full Shot; Eye Level; Static Shot; Centered; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Composite a motionless snail resting on a wet surface onto the provided archive plate featuring a stylized map of Route 66 painted on a station wall. The snail’s tiny presence contrasts with the vast red line of the map. Match the era-locked-look, rainy evening look. Full Shot, Eye Level, Static, Centered, Deep Depth of Field, Wide Angle Lens.
  • negative_prompt: cartoon snail, mismatched lighting, shallow depth of field, moving camera, modern graphics, dry environment
  • reference_summary: A sequence of three images showing a stylized map of Route 66 in the United States, with a red segment progressively moving along the route from Chicago to Los Angeles, indicating progress along the road. The logos and state names remain fixed, while only the position of the red segment changes, …
  • brief notes: composite 4/4 in a row - interleave an archive or synthetic beat so the cut stays easy to follow and cheaper to make; big framing jump (Macro Shot -> Full Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B01
  • beat_summary: The snail takes shelter under a gas station overhang during a sudden, heavy downpour.
  • beat_shift: Exposure -> Cautious safety
  • kind: composite
  • tempo: rapid
  • duration_s: 2.32
  • editorial_note: Tilts to connect her small shelter with the overwhelming weather conditions above.
  • archive_segment_id: F719859_Kilimangiaro_Route_66_La_strada_madre#0123
  • source_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • archive_in: 544.56
  • archive_out: 554.72
  • archive_tc_in: 00:09:04:14
  • archive_tc_out: 00:09:14:18
  • origin_code: cmp
  • notes: Augment archive F157301_DREAMS_ROADS___STREET_OF_AMERICA with AI/VFX (score 0.519) | [4.6 re-select] plate re-selected -> F719859_Kilimangiaro_Route_66_La_strada_madre#0123 (score 0.519)
  • description: Tilting up the dark structure supporting a vibrant neon ‘ROUTE 66’ sign, the snail is composited onto the glowing framework, her silhouette bathed in the stark, colorful contrast of the lively nocturnal lights.
  • archive_summary_en: The sequence shows an arid landscape with rocky hills and a dirt road, initially with an information box about a festival, then without graphics, and finally moves towards an area with a water surface and denser vegetation, always maintaining the mountainous background.
  • revised: True
  • reselection: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0437 -> F719859_Kilimangiaro_Route_66_La_strada_madre#0123

OMC sidecar (001__osc_sc03_0040_comp_gen_a002_v001_kc_snail_puddle.1001.omc.json)

  • asset: osc_sc03_0040_comp_gen_a002_v001_kc_snail_puddle.1001.png
  • deliverable stem: osc_sc03_0040_comp_cmp_a003_v001_kc_snail_puddle.mov
  • era: Kodachrome 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 2121900750
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T11:03:06.238Z
  • generation_seconds: 14
  • task_id: 019f1309-f451-77f0-b4b5-80533e1878f2
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Composite of archive + generated layers.
  • Kitsu preview file id: 8e19e55d-1722-4dec-822f-2356291bee4f
  • Kitsu entity id: 8e4561b3-32b1-46dc-8283-df21ed4def36 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0050

The camera slowly tracks around snail as she gazes up at a glowing circular window, the reflection of a flickering television screen broadcasting the Apollo launch illuminating her face with cosmic awe.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0050_comp_gen_a002_v001_kc_snail_shell.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-03-B02
  • beat_summary: Through the window, the snail watches a television broadcasting the 1969 NASA Apollo mission.
  • beat_shift: mundane survival -> cosmic awe
  • tempo: brisk
  • transition: Eyeline Match

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Low Angle
  • camera_movement: Arc Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 77
  • nb_frames: 77

Continuity

  • big framing jump (Full Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 3.09
  • reference_archive_id: T08340_222_TG2_SI_VIAGGIARE_2008__2009_Viaggio_lungo_la_Route_66
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Close-Up Shot; Low Angle; Arc Shot; Rule of Thirds; Single Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Close-up shot from a low angle slowly arcing around a snail gazing up. The snail’s face is illuminated by the cosmic, flickering glow of a circular window reflecting a 1969 television broadcast of the Apollo launch. The background features dark, atmospheric textures of a rainy evening gas station. Shallow Depth of Field, Medium Lens, awe-struck mood.
  • negative_prompt: static camera, deep depth of field, daylight, modern television, clear urban skyline, dry environment, wide angle
  • reference_summary: The sequence begins with a panoramic view of an urban context characterized by skyscrapers and classic architecture, continues with a slight variation in angle that emphasizes the central building, and ends with a close-up of a dark surface with a circular window reflecting the building seen …
  • brief notes: big framing jump (Full Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B01
  • beat_summary: The snail takes shelter under a gas station overhang during a sudden, heavy downpour.
  • beat_shift: Exposure -> Cautious safety
  • kind: composite
  • tempo: rapid
  • duration_s: 2.32
  • editorial_note: Pans across the gas station to establish the desolate, rainy environment.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0022
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 117.48
  • archive_out: 125.48
  • archive_tc_in: 00:01:57:12
  • archive_tc_out: 00:02:05:12
  • origin_code: cmp
  • notes: Augment archive F157301_DREAMS_ROADS___STREET_OF_AMERICA with AI/VFX (score 1.0066) | [4.6 re-select] plate re-selected -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0022 (score 0.5)
  • description: Composited onto the dusty shoulder of the arid road, the snail braces herself as the camera pans to follow a dark green bus roaring past, leaving her behind as the view opens up to the vast, empty landscape.
  • archive_summary_en: A sequence showing an old beige car with a white container on the roof, marked with the number ‘16596’, as it moves on a paved road. The sequence shifts from a front view to a side view and finally to a rear view, highlighting the vehicle’s details, including the metal cylinders on the back.
  • revised: True
  • reselection: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0412 -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0022

OMC sidecar (001__osc_sc03_0050_comp_gen_a002_v001_kc_snail_shell.1001.omc.json)

  • asset: osc_sc03_0050_comp_gen_a002_v001_kc_snail_shell.1001.png
  • deliverable stem: osc_sc03_0050_comp_cmp_a003_v001_kc_snail_shell.mov
  • era: era-locked-look 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1281327059
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T11:03:49.390Z
  • generation_seconds: 11
  • task_id: 019f130b-e845-7d20-8fea-2fd69cf209a0
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 582bab11-4c04-4a2f-8428-1b3185a83b5d
  • Kitsu entity id: a6d87b7d-9b7c-474e-b0f5-4b5e7347a60d (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0060

A black-and-white broadcast on the television screen shows a bustling city crowd, the camera pushing in on the human faces to emphasize the collective wonder of a species looking toward the stars, mirroring snail’s own silent observation.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0060_comp_gen_a002_v001_kc_snail_rainbow.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-03-B02
  • beat_summary: Through the window, the snail watches a television broadcasting the 1969 NASA Apollo mission.
  • beat_shift: mundane survival -> cosmic awe
  • tempo: brisk
  • transition: Cut

Camera

  • shot_size: Medium Close-Up Shot
  • camera_angle: Eye Level
  • camera_movement: Push In
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 77
  • nb_frames: 77

Archive source

  • archive_id: T95152-221_ALMANACCO_1995#0063
  • archive_tc_in: 00:17:58:23
  • archive_tc_out: 1078.92

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 3.09
  • reference_archive_id: T95152-221_ALMANACCO_1995
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Medium Close-Up Shot; Eye Level; Push In; Centered; Group Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip T95152-221_ALMANACCO_1995 to the cut. Regrade to a 1969 black-and-white television broadcast look, focusing on the medium close-up push-in on the bustling city crowd’s faces. Retime to 3.09 seconds. No generation.
  • reference_summary: A black and white sequence showing a crowd of people moving along a city street, with a vertically striped building in the background. The shot progressively moves from the center of the crowd to a figure in the foreground, suggesting an approaching movement or a transition of focus.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B01
  • beat_summary: The snail takes shelter under a gas station overhang during a sudden, heavy downpour.
  • beat_shift: Exposure -> Cautious safety
  • kind: composite
  • tempo: rapid
  • duration_s: 2.32
  • editorial_note: Returns to a tight close-up, grounding the viewer back in her immediate experience.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0021
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 107.64
  • archive_out: 117.44
  • archive_tc_in: 00:01:47:16
  • archive_tc_out: 00:01:57:11
  • origin_code: cmp
  • notes: Augment archive T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II with AI/VFX (score 1.0066) | [4.6 re-select] plate re-selected -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0021 (score 0.5)
  • description: In a surreal juxtaposition, the snail is composited onto the interior glass of a space capsule, her tiny form silhouetted against the Earth outside as the astronaut slowly raises a gloved hand in the background.
  • archive_summary_en: The sequence shows a person standing next to a white vehicle, who moves slightly to the right and then exits the frame, leaving only the vehicle and the background with other vehicles and trees in view.
  • revised: True
  • reselection: T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II#0120 -> F157301_DREAMS_ROADS___STREET_OF_AMERICA#0021

OMC sidecar (001__osc_sc03_0060_comp_gen_a002_v001_kc_snail_rainbow.1001.omc.json)

  • asset: osc_sc03_0060_comp_gen_a002_v001_kc_snail_rainbow.1001.png
  • deliverable stem: osc_sc03_0060_comp_cmp_a003_v001_kc_snail_rainbow.mov
  • era: era-locked-look 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 570462186
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T11:08:31.464Z
  • generation_seconds: 13
  • task_id: 019f130f-c32b-7513-9a63-2aa65796bdff
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 7d187632-2581-4aa5-b5e3-644b9b0db77e
  • Kitsu entity id: bde42e3b-c5c5-4880-bbc2-1b740834ea7a (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0070

The television broadcast cuts to a vast, distorted desert landscape along route-66, juxtaposing the earthly isolation of the historic highway with the monumental leap into space happening on the same screen.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0070_comp_gen_a002_v001_kc_rainbow_wide.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-03-B02
  • beat_summary: Through the window, snail watches a television broadcasting the 1969 NASA Apollo mission.
  • beat_shift: mundane survival -> cosmic awe
  • tempo: brisk
  • transition: Cross Dissolve

Camera

  • shot_size: Extreme Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 77
  • nb_frames: 77

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0281
  • archive_tc_in: 00:24:45:13
  • archive_tc_out: 1485.52

Continuity

  • monochrome<->colour cut - bridge with a dissolve or regrade so the join doesn’t jar; big framing jump (Medium Close-Up Shot -> Extreme Wide Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 3.09
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Extreme Wide Shot; Eye Level; Static Shot; Rule of Thirds; Point-of-View Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a 1969 television broadcast look, focusing on the extreme wide static shot of the distorted desert landscape along Route 66. Retime to 3.09 seconds. No generation.
  • reference_summary: The sequence begins with a vast desert landscape with a bridge in the distance, moves to a historical sign indicating one of the oldest trading posts on Route 66, and ends with a distorted and blurred vision of the desert, suggesting a journey or transition through an arid and isolated environment.
  • brief notes: monochrome<->colour cut - bridge with a dissolve or regrade so the join doesn’t jar; big framing jump (Medium Close-Up Shot -> Extreme Wide Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B02
  • beat_summary: A television inside broadcasts the crackling audio of the 1969 NASA moon landing.
  • beat_shift: Mundane survival -> Cosmic awareness
  • kind: archive
  • tempo: rapid
  • duration_s: 1.74
  • editorial_note: Follows her attention to the television broadcasting the historic moon landing.
  • archive_segment_id: C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO#0121
  • source_archive_id: C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO
  • archive_in: 1035.6
  • archive_out: 1038.96
  • archive_tc_in: 00:17:15:15
  • archive_tc_out: 00:17:18:24
  • origin_code: arc
  • notes: Strong archive match (score 1.0182); relations: conceptual, lexical, narrative | [4.6 re-select] plate re-selected -> C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO#0121 (score 0.5)
  • description: A static, black-and-white broadcast view captures the historic moment of two astronauts standing on the lunar surface beside their module, the stark contrast emphasizing the monumental leap of human technology.
  • archive_summary_en: A sequence of a crowded street in a historical context, with people in formal attire and vintage cars. The image quality progressively degrades, with visual artifacts increasing towards the end, making the right side of the image almost unreadable.
  • revised: True
  • reselection: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1#0014 -> C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO#0121

OMC sidecar (001__osc_sc03_0070_comp_gen_a002_v001_kc_rainbow_wide.1001.omc.json)

  • asset: osc_sc03_0070_comp_gen_a002_v001_kc_rainbow_wide.1001.png
  • deliverable stem: osc_sc03_0070_comp_cmp_a003_v001_kc_rainbow_wide.mov
  • era: era-locked-look 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 2141988950
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T11:16:50.105Z
  • generation_seconds: 11
  • task_id: 019f1310-e25b-7c22-9b9f-0b069ae9e3e4
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 9f9bfbbc-9fa6-4056-b59a-f2f04437f3c6
  • Kitsu entity id: 70d51dcf-da2e-4cf3-b127-c607614a1a6f (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0080

A sweeping pan across a rocky, rain-washed landscape reveals a sudden burst of lush vegetation and a shimmering pool of water, as vibrant era-locked-look colors flood the scene, signaling a transcendent awakening after the storm.

Preview

Latest Kitsu preview (revision 1, source: osc_sc03_0080_comp_gen_a002_v001_kc_color_arrives.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-03-B03
  • beat_summary: The rain stops, a rainbow appears, and vibrant Kodachrome color bleeds into the previously black-and-white world.
  • beat_shift: stark reality -> vibrant awakening
  • tempo: brisk
  • transition: Match Cut

Camera

  • shot_size: Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Pan
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 77
  • nb_frames: 77

Archive source

  • archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre#0123
  • archive_tc_in: 00:09:04:14
  • archive_tc_out: 544.56

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 3.09
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Wide Shot; Eye Level; Pan; Rule of Thirds; Point-of-View Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip F719859_Kilimangiaro_Route_66_La_strada_madre to the cut. Regrade to a vibrant 1969 Kodachrome look, focusing on the wide panning shot across the rocky landscape revealing lush vegetation and a shimmering pool of water. Retime to 3.09 seconds. No generation.
  • reference_summary: The sequence shows an arid landscape with rocky hills and a dirt road, initially with an information box about a festival, then without graphics, and finally moves towards an area with a water surface and denser vegetation, always maintaining the mountainous background.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B02
  • beat_summary: A television inside broadcasts the crackling audio of the 1969 NASA moon landing.
  • beat_shift: Mundane survival -> Cosmic awareness
  • kind: archive
  • tempo: rapid
  • duration_s: 1.74
  • editorial_note: Tracks wide to show the glowing broadcast contrasting with the dark, rainy station.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0349
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 1874.36
  • archive_out: 1879.32
  • archive_tc_in: 00:31:14:09
  • archive_tc_out: 00:31:19:08
  • origin_code: arc
  • notes: Strong archive match (score 1.0); relations: conceptual, narrative
  • description: Tracking smoothly alongside a white car as it cuts through the arid desert, the wide framing captures the fluid, solitary motion of human travel across the vast, open territory.
  • archive_summary_en: The sequence shows an arid landscape that transforms into a travel scene with a white car driving along a road in a desert environment. The environmental continuity and the progression of the vehicle suggest a fluid movement through an open and solitary territory.

OMC sidecar (001__osc_sc03_0080_comp_gen_a002_v001_kc_color_arrives.1001.omc.json)

  • asset: osc_sc03_0080_comp_gen_a002_v001_kc_color_arrives.1001.png
  • deliverable stem: osc_sc03_0080_comp_cmp_a003_v001_kc_color_arrives.mov
  • era: Kodachrome 1969
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 716567449
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T11:11:10.502Z
  • generation_seconds: 14
  • task_id: 019f1313-167e-76f0-9353-19773f8766c7
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 9e809421-caba-4466-9f99-0e43aa26170a
  • Kitsu entity id: 517a9d9f-3c47-4ec4-997f-bb43712d4acb (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0090

Pushing in from a wide view of a vibrant, neon-lit route-66 diner interior to a close-up of a hanging toy car, the rich, saturated colors of the memorabilia celebrate the vibrant reality that emerges as the rain clears.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-03-B03
  • beat_summary: The rain stops, a rainbow appears, and vibrant era-locked-look color bleeds into the previously black-and-white world.
  • beat_shift: stark reality -> vibrant awakening
  • tempo: brisk
  • transition: Cut

Camera

  • shot_size: Medium Shot
  • camera_angle: Eye Level
  • camera_movement: Push In
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 77
  • nb_frames: 77

Archive source

  • archive_id: Roadbook#0185
  • archive_tc_in: 00:13:57:13
  • archive_tc_out: 837.52

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 3.09
  • reference_archive_id: Roadbook
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Medium Shot; Eye Level; Push In; Centered; Point-of-View Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip Roadbook to the cut. Regrade to a rich, saturated 1969 Kodachrome look, focusing on the medium shot pushing in on the neon-lit diner interior and hanging toy car. Retime to 3.09 seconds. No generation.
  • reference_summary: A sequence of three shots showing the interior of a Route 66 themed venue, with vintage decorations, counter, refrigerated display cases and memorabilia, moving from a wide view to a closer one and finally to a detail of the toy car hanging above the display case.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B02
  • beat_summary: A television inside broadcasts the crackling audio of the 1969 NASA moon landing.
  • beat_shift: Mundane survival -> Cosmic awareness
  • kind: archive
  • tempo: rapid
  • duration_s: 1.74
  • editorial_note: Pans across the screen to emphasize the crackling audio of human achievement.
  • archive_segment_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1#0014
  • source_archive_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1
  • archive_in: 68.96
  • archive_out: 81.84
  • archive_tc_in: 00:01:08:24
  • archive_tc_out: 00:01:21:21
  • origin_code: arc
  • notes: Strong archive match (score 1.0); relations: conceptual, narrative | [4.6 re-select] plate re-selected -> T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1#0014 (score 0.516)
  • description: Panning from an astronaut working inside the cabin to the breathtaking curvature of the Earth visible through a circular window, the shot visually connects human endeavor with the vastness of space.
  • archive_summary_en: The sequence shows three moments related to the lunar landing: a close-up of Neil Armstrong in an indoor environment, a close-up of Buzz Aldrin with a camera, and finally an outdoor scene with both astronauts on the lunar surface next to the lunar module. The images are in black and white and show …
  • revised: True
  • reselection: T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II#0125 -> T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1#0014

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu entity id: e1356ebf-5ad6-4818-882f-5b62d8c0b44c (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc03_0100

A glowing neon sign on a city building bursts into brilliant color as a vehicle passes, the radiant light cutting through the lingering dampness before dissolving into a stylized, dreamlike distortion that heightens the emotional climax.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc03
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-03-B03
  • beat_summary: The rain stops, a rainbow appears, and vibrant era-locked-look color bleeds into the previously black-and-white world.
  • beat_shift: stark reality -> vibrant awakening
  • tempo: brisk
  • transition: Cut

Camera

  • shot_size: Medium Wide Shot
  • camera_angle: Low Angle
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 77
  • nb_frames: 77

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0026
  • archive_tc_in: 00:02:21:09
  • archive_tc_out: 141.36

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. GAS STATION, ROUTE 66 - LATE AFTERNOON INTO EVENING
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 3.09
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1969. mood: Atmospheric, Transformative, Awe-struck. time of day: late afternoon into evening.
  • camera: Medium Wide Shot; Low Angle; Static Shot; Rule of Thirds; Point-of-View Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a 1969 atmospheric evening look, focusing on the medium wide low-angle static shot of the glowing neon sign distorting into a dreamlike state. Retime to 3.09 seconds. No generation.
  • reference_summary: The sequence shows a luminous ‘Lo Mith’s’ sign on a city building, initially partially hidden by a moving vehicle, then completely visible, and finally distorted by strong digital noise that compromises its clarity.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-03-B02
  • beat_summary: A television inside broadcasts the crackling audio of the 1969 NASA moon landing.
  • beat_shift: Mundane survival -> Cosmic awareness
  • kind: archive
  • tempo: rapid
  • duration_s: 1.74
  • editorial_note: Pulls out to an extreme wide, juxtaposing the monumental broadcast with the tiny viewer.
  • archive_segment_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1#0061
  • source_archive_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1
  • archive_in: 622.12
  • archive_out: 632.24
  • archive_tc_in: 00:10:22:03
  • archive_tc_out: 00:10:32:06
  • origin_code: arc
  • notes: Strong archive match (score 1.0); relations: conceptual, narrative
  • description: Tilting up powerfully as a massive rocket launches into the sky, the telephoto compression emphasizes the sheer scale and explosive energy of humanity’s departure from Earth.
  • archive_summary_en: The sequence begins with a newspaper announcing a lunar event, continues with the launch of a rocket, and ends with an astronaut on the lunar surface, creating a visual narrative of space travel from the moment of the announcement to the landing.

Notes

  • Part of sequence sc03 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu entity id: acd0159c-6ecb-46ed-b3a7-7e7a76aec150 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc04 - 04 The Decay & the Dream

Sequence sc04 - The Decay & the Dream

Preview

First available Kitsu preview for this sequence (from shot sc04_0010). 14 of 22 shots in this sequence have Kitsu previews.

Facts

Shots

  • sc04_0010: A slow tracking shot follows snail as she crawls past the rusted, decaying sign of the ‘Old 66 Cafe’, the faded lettering emphasizing…
  • sc04_0020: The camera tilts up from the snail’s solitary path to reveal a flickering, broken neon sign against the dark sky, its dying light casting…
  • sc04_0030: Pushing in on the snail as she navigates past the rusted shell of an abandoned Airstream trailer, the blurred, overgrown wilderness in the…
  • sc04_0040: A wide, static shot captures the snail as a tiny speck on the cracked pavement, while a weathered yellow pickup truck speeds away into the…
  • sc04_0050: The camera dollies backward ahead of the snail, framing her relentless crawl against the vast, empty expanse of the desert highway,…
  • sc04_0060: A low-angle pan follows the snail as a dark green bus roars past and disappears into the horizon, the sudden emptiness of the frame…
  • sc04_0070: The camera tracks the snail from a high angle as she navigates the cracked, dusty asphalt, her slow, deliberate movements visually echoing…
  • sc04_0080: A static wide shot frames the snail perfectly centered on the endless, empty road stretching toward the horizon, the stark symmetry…
  • sc04_0090: A stark, high-contrast shot of an astronaut walking slowly across the desolate lunar surface, the sterile, dusty environment sharply…
  • sc04_0100 (canceled): A rapid pan follows a white car speeding through the arid desert, the fluid motion of earthly travel contrasting sharply with the clinical…
  • sc04_0110 (canceled): A dynamic tracking shot captures a dark green bus hurtling through the barren landscape, the aggressive earthly momentum setting up a…
  • sc04_0120 (canceled): Pushing in from the complex, technological interior of a spacecraft to the vast, mysterious expanse of the lunar surface outside the…
  • sc04_0130 (canceled): A static shot of an astronaut inside the clinical lunar module cuts to a breathtaking view of the crescent moon, the stark transition…
  • sc04_0140 (canceled): A rapid zoom out from a detailed map of route-66 to the endless, empty desert road visually compresses the earthly journey, contrasting the…
  • sc04_0150 (canceled): A whip pan blurs the snail’s slow crawl past towering, modern wind turbines, the dizzying motion serving as a jarring transition between…
  • sc04_0160 (canceled): A smooth tracking shot from a moving vehicle captures a weathered ‘Old 66 Life’ sign surrounded by encroaching nature, emphasizing the…
  • sc04_0170 (canceled): A static shot holds on a rusted ‘Old 66 Cafe’ sign as digital artifacts slowly distort the image, visually representing the fading memory…
  • sc04_0180 (canceled): A slow tilt up past a glowing, isolated Route 66 neon sign reveals the vast, dark night sky above, bridging the fading earthly artifacts…
  • sc04_0190 (canceled): A wide panning shot follows a solitary yellow pickup truck driving through the vast, empty desert, its small presence dwarfed by the…
  • sc04_0200 (canceled): A static close-up of a decorative Route 66 plaque serves as a melancholic marker of the past, its fixed map of ‘America’s Main Street’…
  • sc04_0210 (canceled): A historical shot of a bustling, formal street scene slowly degrades into heavy visual artifacts, the dissolving image poetically mirroring…
  • sc04_0220 (canceled): The camera slowly cranes up from the snail resting on a patch of weeds breaking through the asphalt, revealing a vast night sky where the…

sc04_0010

A slow tracking shot follows snail as she crawls past the rusted, decaying sign of the ‘Old 66 Cafe’, the faded lettering emphasizing the quiet abandonment of the once-vibrant highway.

Preview

Latest Kitsu preview (revision 1, source: sc04_0010-astrid-emerveillee).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Cross Dissolve

Camera

  • shot_size: Medium Shot
  • camera_angle: Ground Level
  • camera_movement: Tracking Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Shot; Ground Level; Tracking Shot; Leading Lines; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Medium tracking shot at ground level following a snail crawling past the rusted, decaying base of an ‘Old 66 Cafe’ sign. The deep depth of field reveals a quiet, abandoned highway landscape with faded lettering and melancholic, clinical lighting typical of the 1970s-present era. Wide Angle Lens, leading lines.
  • negative_prompt: shallow depth of field, static camera, bright cheerful lighting, pristine signs, modern bustling traffic, digital disturbances on screen
  • reference_summary: The sequence shows two distinct road signs related to route-66: first a cafe sign with the name ‘Old. 66 Cafe’, then an official ‘ROUTE US 66’ sign, which repeats in the third scene with an increase in digital visual disturbances.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Bridges the decades into a rapid montage of decaying infrastructure.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: A rapid whip pan blurs the transition across the arid landscape, revealing a once-busy desert road now cracked and empty, visually accelerating the passage of time and the onset of decay.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 8fbbe488-b0fe-4fa4-bd8f-a1e1ef259b2a
  • Kitsu entity id: 48a01695-b5fe-4308-bd17-0eec83659831 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc04_0020

The camera tilts up from snail’s solitary path to reveal a flickering, broken neon sign against the dark sky, its dying light casting long, melancholic shadows over the weed-choked asphalt.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Cut on Action

Camera

  • shot_size: Full Shot
  • camera_angle: Low Angle
  • camera_movement: Tilt
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Full Shot; Low Angle; Tilt; Rule of Thirds; Single Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Full shot from a low angle tilting up from a snail’s solitary path on weed-choked asphalt to a flickering, broken route-66 neon sign against a dark, melancholic night sky. The dying light casts long shadows. Shallow depth of field, Medium Lens, Rule of Thirds, 1970s-present era aesthetic.
  • negative_prompt: deep depth of field, static camera, bright daylight, pristine neon sign, bustling city, wide angle lens
  • reference_summary: The sequence shows a progression of neon signs against a dark background, starting with the letters ‘EL’, moving to a stylized figure with an arrow and ending with the iconic ‘ROUTE 66’ lettering, all characterized by strong color contrast and a lively night atmosphere.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pushes in to accelerate the visual rhythm as the years fly by.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Pushing in on a rusted, abandoned roadside relic that resembles the metallic starkness of a lunar module, the shallow focus isolates the decaying human artifact against the encroaching desert.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 75b016c0-55ce-424d-8f92-043fca8ea6ce
  • Kitsu entity id: b85d1e9f-d7bd-4472-b001-d0ad26e164b9 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc04_0030

Pushing in on snail as she navigates past the rusted shell of an abandoned Airstream trailer, the blurred, overgrown wilderness in the background underscores the stark transition from human bustling to natural reclamation.

Preview

Latest Kitsu preview (revision 1, source: sc04_0030-insert-patte-renard).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Match Cut

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Eye Level
  • camera_movement: Push In
  • lens: Telephoto Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

Continuity

  • big framing jump (Full Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Close-Up Shot; Eye Level; Push In; Centered; Single Shot; Bokeh; Telephoto Lens; Frame Rate
  • prompt: Close-up shot pushing in on a snail navigating past the rusted, abandoned shell of an Airstream trailer. The background is a heavy bokeh of overgrown wilderness, emphasizing natural reclamation and a melancholic, clinical mood. Centered composition, Telephoto Lens, 1970s-present era textures.
  • negative_prompt: deep depth of field, wide angle, static camera, pristine Airstream, human figures, visual defects on screen, bright cheerful lighting
  • reference_summary: The sequence shows a progression from an Airstream caravan in an open environment, with people entering and exiting, to a more remote and wild scene, where human figures are present in a blurred natural context, with evident visual defects in the third scene.
  • brief notes: big framing jump (Full Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Quick close-up to maintain the frantic pace of the montage.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: A static, low-angle close-up of a faded, bullet-riddled ‘ROUTE US 66’ sign, its peeling paint standing in stark contrast to the blurred, desolate sky behind it.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: d4627a2a-20a6-4c92-a3d3-1992f2700493
  • Kitsu entity id: e733ae43-c382-4858-baf3-048c8b0d6a37 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc04_0040

A wide, static shot captures snail as a tiny speck on the cracked pavement, while a weathered yellow pickup truck speeds away into the barren desert distance, leaving her entirely alone on the empty road.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Smash Cut

Camera

  • shot_size: Extreme Wide Shot
  • camera_angle: High Angle
  • camera_movement: Static Shot
  • lens: Ultra Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

Continuity

  • big framing jump (Close-Up Shot -> Extreme Wide Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Extreme Wide Shot; High Angle; Static Shot; Rule of Thirds; Single Shot; Deep Depth of Field; Ultra Wide Angle Lens; Frame Rate
  • prompt: Extreme wide static shot from a high angle capturing a tiny snail on cracked pavement. In the deep depth of field, a weathered yellow pickup truck speeds away into a barren, melancholic desert distance, leaving the snail entirely alone. Ultra Wide Angle Lens, Rule of Thirds, 1970s-present era.
  • negative_prompt: shallow depth of field, moving camera, close-up, lush green landscape, modern sleek vehicles, bright cheerful lighting
  • reference_summary: A sequence of three shots showing a yellow pickup truck with ‘Route US 66’ decorations moving on a road in a desert environment. The camera follows the vehicle from different angles: front, side and rear, highlighting the driver and the landscape context.
  • brief notes: big framing jump (Close-Up Shot -> Extreme Wide Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Tracks the continuous decay, keeping the eye moving seamlessly.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Tracking tightly alongside the snail at ground level, the macro lens blurs the crumbling asphalt into a chaotic texture as she steadily bypasses the fallen debris of a forgotten era.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 582a145f-b4b7-4f8f-b449-0c14cc6dc458
  • Kitsu entity id: 9161a382-070b-42ac-b225-b6ac17b140e7 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc04_0050

The camera dollies backward ahead of snail, framing her relentless crawl against the vast, empty expanse of the desert highway, emphasizing the sheer scale of the solitary journey that remains.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Full Shot
  • camera_angle: Ground Level
  • camera_movement: Dolly Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Full Shot; Ground Level; Dolly Shot; Leading Lines; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Full shot at ground level dollying backward ahead of a snail crawling relentlessly along a vast, empty desert highway. The deep depth of field showcases the sheer scale of the solitary, cracked road stretching into an arid landscape. Wide Angle Lens, leading lines, melancholic 1970s-present era lighting.
  • negative_prompt: shallow depth of field, static camera, close-up, bustling traffic, lush vegetation, white car on screen
  • reference_summary: The sequence shows an arid landscape that transforms into a travel scene with a white car traveling along a road in a desert environment. The environmental continuity and the progression of the vehicle suggest a fluid movement through an open and solitary territory.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Tilts to reveal more of the crumbling highway in rapid succession.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Tilting down from the harsh, empty sky to the fractured pavement, the wide shot emphasizes the snail’s isolation amidst the sprawling, decaying infrastructure of the abandoned highway.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 281abfce-5c9a-4121-80e9-8305f8f5950d
  • Kitsu entity id: 0bb5645f-d0ba-41b4-8fe1-56343d2caa04 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc04_0060

A low-angle pan follows snail as a dark green bus roars past and disappears into the horizon, the sudden emptiness of the frame highlighting the fading presence of humanity on the historic route.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Cut on Action

Camera

  • shot_size: Medium Wide Shot
  • camera_angle: Low Angle
  • camera_movement: Pan
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Wide Shot; Low Angle; Pan; Rule of Thirds; Single Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Medium wide low-angle pan following a snail as a dark green bus roars past in the deep depth of field and disappears into the horizon of a barren, arid landscape. The sudden emptiness highlights a melancholic, clinical mood. Medium Lens, Rule of Thirds, 1970s-present era.
  • negative_prompt: shallow depth of field, static camera, high angle, bustling city, modern sleek bus, bright cheerful lighting
  • reference_summary: A sequence of three scenes showing a dark green bus traveling on a road in an arid area, initially approaching, then moving fast and finally out of frame, with the camera passing it and moving towards an open view of the road and the surrounding landscape.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pans quickly across the empty road to emphasize abandonment.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Panning quickly past the skeletal remains of a collapsed billboard, the camera catches the snail’s relentless forward motion, contrasting her enduring life with the dead metal around her.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 087bb074-5591-4173-a3c6-a599440e0ccd
  • Kitsu entity id: d9ea8cda-a9d6-44a6-ae3b-d1bf6c52a4b7 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:52.

sc04_0070

The camera tracks snail from a high angle as she navigates the cracked, dusty asphalt, her slow, deliberate movements visually echoing the desolate, low-gravity steps of an astronaut on a barren lunar surface.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Medium Shot
  • camera_angle: High Angle
  • camera_movement: Tracking Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Shot; High Angle; Tracking Shot; Centered; Single Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Medium high-angle tracking shot following a snail navigating cracked, dusty asphalt. The stark, high-contrast lighting and desolate textures visually echo a barren lunar surface, creating a melancholic, clinical atmosphere. Centered composition, Deep Depth of Field, Medium Lens, 1970s-present era.
  • negative_prompt: shallow depth of field, static camera, low angle, lush vegetation, astronaut on screen, bright cheerful lighting
  • reference_summary: A sequence showing an astronaut in a spacesuit walking slowly on the lunar surface, with a desolate landscape and a dark shadow in the background. The sunlight creates sharp contrasts, and the astronaut moves progressively across the dusty and rocky terrain.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Drops to a drone shot for a sudden, vast perspective of the decay.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: A drone shot looking straight down reveals the symmetrical, fading painted lines of the highway, now broken by weeds, with the tiny snail acting as the only moving element in the vast decay.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 258f7d50-ec3f-4977-ac07-ddcf6239e92b
  • Kitsu entity id: 3be18501-e0e6-40d5-a047-34ef7c18977a (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0080

A static wide shot frames snail perfectly centered on the endless, empty road stretching toward the horizon, the stark symmetry emphasizing her isolation in the vast, decaying landscape.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B01
  • beat_summary: The snail continues past boarded-up diners and weed-choked gas stations as the road empties.
  • beat_shift: vibrant era -> quiet abandonment
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Wide Shot; Eye Level; Static Shot; Symmetry; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Wide static shot at eye level framing a snail perfectly centered on an endless, empty desert road stretching toward the horizon. The stark symmetry and deep depth of field emphasize isolation in a vast, decaying landscape. Wide Angle Lens, melancholic 1970s-present era lighting.
  • negative_prompt: shallow depth of field, moving camera, off-center composition, bustling traffic, text overlays, map graphics on screen
  • reference_summary: The sequence begins with a view of a deserted road in the desert with textual and graphic overlays of route-66, moves to a detailed map of the road’s route across the United States, and ends with a frontal view of an empty road stretching towards the horizon in an arid landscape.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pushes in on the ruins to keep the montage’s forward momentum.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Pushing in low along a deep crack in the asphalt that serves as a leading line, the shallow focus highlights the snail’s determined face as she navigates the ruined terrain.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: d1199d98-d4f2-4d9c-9121-684b51fa1d74
  • Kitsu entity id: 6b7edce5-0e1a-4eb0-80bd-fb68d60d8373 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0090

A stark, high-contrast shot of an astronaut walking slowly across the desolate lunar surface, the sterile, dusty environment sharply juxtaposing the organic decay of the earthly highway.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0090_comp_gen_a002_v001_vh_vhs_diner_low.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: no
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B02
  • beat_summary: The decaying earthly landscape is sharply intercut with the pristine, clinical interior of a space capsule in lunar orbit.
  • beat_shift: earthly decay -> sterile progress
  • tempo: rapid
  • transition: Smash Cut

Camera

  • shot_size: Full Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1#0065
  • archive_tc_in: 00:10:44:16
  • archive_tc_out: 644.64

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Full Shot; Eye Level; Static Shot; Rule of Thirds; Single Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1 to the cut. Regrade to a stark, high-contrast melancholic look, focusing on the full static shot of the astronaut walking on the lunar surface. Retime to 1.41 seconds. No generation.
  • reference_summary: A sequence showing an astronaut in a spacesuit walking slowly on the lunar surface, with a desolate landscape and a dark shadow in the background. The sunlight creates sharp contrasts, and the astronaut moves progressively across the dusty and rocky terrain.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B01
  • beat_summary: The highway infrastructure decays and empties as years pass in a rapid montage.
  • beat_shift: Bustling life -> Quiet abandonment
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Another rapid close-up to heighten the sense of fleeting time.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: A static Dutch angle creates a sense of unease, framing a rusted hubcap half-buried in the dirt while snail slowly crawls over its jagged, oxidized edge.

OMC sidecar (001__osc_sc04_0090_comp_gen_a002_v001_vh_vhs_diner_low.1001.omc.json)

  • asset: osc_sc04_0090_comp_gen_a002_v001_vh_vhs_diner_low.1001.png
  • deliverable stem: osc_sc04_0090_comp_cmp_a003_v001_vh_vhs_diner_low.mov
  • era: era-locked-look 1980s-90s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1235545422
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:00:56.031Z
  • generation_seconds: 14
  • task_id: 019f1340-a8eb-7a70-9f68-194da0d5fa8b
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: b40a58af-e9ab-461a-8c1e-a1e9a5fd32a7
  • Kitsu entity id: 47c2b612-2c6f-48ac-893c-33e2e2ce4c27 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0100

A rapid pan follows a white car speeding through the arid desert, the fluid motion of earthly travel contrasting sharply with the clinical precision of the space age.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0100_comp_gen_a002_v001_vh_vhs_diner_flat.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B02
  • beat_summary: The decaying earthly landscape is sharply intercut with the pristine, clinical interior of a space capsule in lunar orbit.
  • beat_shift: earthly decay -> sterile progress
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Pan
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0349
  • archive_tc_in: 00:31:14:09
  • archive_tc_out: 1874.36

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Wide Shot; Eye Level; Pan; Leading Lines; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a clinical, accelerated look, focusing on the wide panning shot of the white car speeding through the arid desert. Retime to 1.41 seconds. No generation.
  • reference_summary: The sequence shows an arid landscape that transforms into a travel scene with a white car traveling along a road in a desert environment. The environmental continuity and the progression of the vehicle suggest a fluid movement through an open and solitary territory.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pedestals up to show nature beginning its reclamation of the asphalt.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Pedestaling up from the overgrown weeds bursting through the cracked asphalt, the wide shot reveals the snail pressing onward as nature aggressively reclaims the forgotten human road.

OMC sidecar (001__osc_sc04_0100_comp_gen_a002_v001_vh_vhs_diner_flat.1001.omc.json)

  • asset: osc_sc04_0100_comp_gen_a002_v001_vh_vhs_diner_flat.1001.png
  • deliverable stem: osc_sc04_0100_comp_cmp_a003_v001_vh_vhs_diner_flat.mov
  • era: era-locked-look 1980s-90s
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 933379239
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:04:29.311Z
  • generation_seconds: 18
  • task_id: 019f1343-5d3a-7283-897a-0191b8820096
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: e8b0d0fc-4f2f-4996-92a8-18506c1c3ff7
  • Kitsu entity id: 4178b3eb-548d-49a2-91f5-493e408c00d3 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0110

A dynamic tracking shot captures a dark green bus hurtling through the barren landscape, the aggressive earthly momentum setting up a jarring cut to the silent vacuum of space.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0110_comp_gen_a002_v001_sp_capsule_interior.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B02
  • beat_summary: The decaying earthly landscape is sharply intercut with the pristine, clinical interior of a space capsule in lunar orbit.
  • beat_shift: earthly decay -> sterile progress
  • tempo: rapid
  • transition: Cut on Action

Camera

  • shot_size: Medium Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Tracking Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0412
  • archive_tc_in: 00:37:15:18
  • archive_tc_out: 2235.72

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Wide Shot; Eye Level; Tracking Shot; Rule of Thirds; Single Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a clinical, accelerated look, focusing on the medium wide tracking shot of the dark green bus hurtling through the barren landscape. Retime to 1.41 seconds. No generation.
  • reference_summary: A sequence of three scenes showing a dark green bus traveling on a road in an arid area, initially approaching, then moving fast and finally out of frame, with the camera passing it and moving towards an open view of the road and the surrounding landscape.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Tracks the snail to remind the viewer of her constant presence amidst rapid change.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Tracking smoothly with the snail, the shallow depth of field turns the encroaching desert flora into a soft, green blur, emphasizing her peaceful integration into the reclaiming wilderness.

OMC sidecar (001__osc_sc04_0110_comp_gen_a002_v001_sp_capsule_interior.1001.omc.json)

  • asset: osc_sc04_0110_comp_gen_a002_v001_sp_capsule_interior.1001.png
  • deliverable stem: osc_sc04_0110_comp_cmp_a003_v001_sp_capsule_interior.mov
  • era: Space
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1830234183
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:15:57.315Z
  • generation_seconds: 28
  • task_id: 019f134e-2b8a-7580-a216-930fdc583747
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 5393621a-4791-4ea6-9dfe-1858304e3410
  • Kitsu entity id: 31f397e0-56ad-4033-8524-127d50f23470 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0120

Pushing in from the complex, technological interior of a spacecraft to the vast, mysterious expanse of the lunar surface outside the window, the shot emphasizes the sterile progress of humanity leaving Earth behind.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0120_comp_gen_a002_v001_sp_nutella_reveal.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B02
  • beat_summary: The decaying earthly landscape is sharply intercut with the pristine, clinical interior of a space capsule in lunar orbit.
  • beat_shift: earthly decay -> sterile progress
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Medium Shot
  • camera_angle: Eye Level
  • camera_movement: Push In
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II#0150
  • archive_tc_in: 00:11:46:24
  • archive_tc_out: 706.96

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Shot; Eye Level; Push In; Centered; Point-of-View Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II to the cut. Regrade to a sterile, melancholic look, focusing on the medium shot pushing in from the spacecraft interior to the lunar surface outside the window. Retime to 1.41 seconds. No generation.
  • reference_summary: The sequence shows a progression from a technological internal setting of a spacecraft towards an external view of a lunar or extraterrestrial surface, culminating in a night scene with a distant luminous object. The focus shifts from internal control to external exploration, with an increase in …

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Macro shot of the cracked road to contrast her scale with the decaying world.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: A static macro shot captures a delicate desert flower blooming from a fissure in the road, with the snail gently gliding past it, highlighting the quiet victory of nature over concrete.

OMC sidecar (001__osc_sc04_0120_comp_gen_a002_v001_sp_nutella_reveal.1001.omc.json)

  • asset: osc_sc04_0120_comp_gen_a002_v001_sp_nutella_reveal.1001.png
  • deliverable stem: osc_sc04_0120_comp_cmp_a003_v001_sp_nutella_reveal.mov
  • era: Space
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 850652361
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:17:59.311Z
  • generation_seconds: 15
  • task_id: 019f1350-2eff-76b2-8129-d2aa97fb1f5e
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: e70f3091-295f-47a2-bfd5-91e091e22f62
  • Kitsu entity id: 23c57e74-3857-4604-859b-e525a4794421 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0130

A static shot of an astronaut inside the clinical lunar module cuts to a breathtaking view of the crescent moon, the stark transition highlighting the ultimate detachment from the decaying world below.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0130_comp_gen_a002_v001_sp_instrument_panel.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B02
  • beat_summary: The decaying earthly landscape is sharply intercut with the pristine, clinical interior of a space capsule in lunar orbit.
  • beat_shift: earthly decay -> sterile progress
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Medium Close-Up Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II#0180
  • archive_tc_in: 00:15:34:05
  • archive_tc_out: 934.2

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Close-Up Shot; Eye Level; Static Shot; Rule of Thirds; Single Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip T09143_211_Tg3_Pomeriggio_Missione_spaziale_Artemis_II to the cut. Regrade to a clinical, melancholic look, focusing on the medium close-up static shot transitioning from the astronaut to the crescent moon. Retime to 1.41 seconds. No generation.
  • reference_summary: The sequence begins with an astronaut inside a lunar module, then moves to a panoramic view of the crescent moon from the lunar soil, ending with a black screen. The progression suggests a passage from the closed and technological environment of the spacecraft to the immensity and solitude of …

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pans across the overgrowth, continuing the relentless rhythm.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Panning down from a sun-bleached, vine-covered road sign to the cracked pavement below, the high angle frames the snail moving steadily into the open leading room of the reclaimed highway.

OMC sidecar (001__osc_sc04_0130_comp_gen_a002_v001_sp_instrument_panel.1001.omc.json)

  • asset: osc_sc04_0130_comp_gen_a002_v001_sp_instrument_panel.1001.png
  • deliverable stem: osc_sc04_0130_comp_cmp_a003_v001_sp_instrument_panel.mov
  • era: Space
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1155639350
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:18:39.135Z
  • generation_seconds: 15
  • task_id: 019f1350-acd3-77a3-9d0c-1d7293d1605e
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 90f0dc7d-a546-4c77-9eb2-cdd85e51be11
  • Kitsu entity id: 59b3c32d-8ba7-4d56-9d7a-ef7010ec9502 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0140

A rapid zoom out from a detailed map of route-66 to the endless, empty desert road visually compresses the earthly journey, contrasting the grounded decay with the boundless reach of the space age.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0140_comp_gen_a002_v001_sp_capsule_roll.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B02
  • beat_summary: The decaying earthly landscape is sharply intercut with the pristine, clinical interior of a space capsule in lunar orbit.
  • beat_shift: earthly decay -> sterile progress
  • tempo: rapid
  • transition: Jump Cut

Camera

  • shot_size: Wide Shot
  • camera_angle: High Angle
  • camera_movement: Zoom Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0406
  • archive_tc_in: 00:36:44:14
  • archive_tc_out: 2204.56

Continuity

  • big framing jump (Medium Close-Up Shot -> Wide Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Wide Shot; High Angle; Zoom Shot; Symmetry; Point-of-View Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a clinical, accelerated look, focusing on the wide high-angle zoom out from the Route 66 map to the empty desert road. Retime to 1.41 seconds. No generation.
  • reference_summary: The sequence begins with a view of a deserted road in the desert with textual and graphic overlays of Route 66, moves to a detailed map of the road’s route across the United States, and ends with a frontal view of an empty road stretching towards the horizon in an arid landscape.
  • brief notes: big framing jump (Medium Close-Up Shot -> Wide Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pushes in on the natural elements overtaking the human-made structures.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Pushing in slowly along the crumbling yellow center line, the deep focus captures the snail’s solitary trek as the road ahead is swallowed by drifting sand and resilient desert scrub.

OMC sidecar (001__osc_sc04_0140_comp_gen_a002_v001_sp_capsule_roll.1001.omc.json)

  • asset: osc_sc04_0140_comp_gen_a002_v001_sp_capsule_roll.1001.png
  • deliverable stem: osc_sc04_0140_comp_cmp_a003_v001_sp_capsule_roll.mov
  • era: Space
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 985720728
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:19:42.612Z
  • generation_seconds: 27
  • task_id: 019f1351-9713-7bc2-a8e1-b79da08d20a2
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: bb40fb96-1fab-4416-a8b7-9f50c8841722
  • Kitsu entity id: d5610882-fd0b-4c09-9027-e9090b94e3b9 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0150

A whip pan blurs snail’s slow crawl past towering, modern wind turbines, the dizzying motion serving as a jarring transition between the grounded reality of the highway and the high-tech realm of space.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0150_comp_gen_a002_v001_sp_earth_porthole.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B02
  • beat_summary: The decaying earthly landscape is sharply intercut with the pristine, clinical interior of a space capsule in lunar orbit.
  • beat_shift: earthly decay -> sterile progress
  • tempo: rapid
  • transition: Smash Cut

Camera

  • shot_size: Medium Shot
  • camera_angle: Low Angle
  • camera_movement: Whip Pan
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Shot; Low Angle; Whip Pan; Dynamic Symmetry; Single Shot; Shallow Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Medium low-angle whip pan blurring past a snail’s slow crawl to reveal towering, modern wind turbines in an arid landscape. The dizzying motion creates a jarring, clinical transition. Shallow depth of field, Wide Angle Lens, dynamic symmetry, 1970s-present era lighting.
  • negative_prompt: static camera, deep depth of field, high angle, lush green landscape, distortion on screen, bright cheerful lighting
  • reference_summary: The sequence begins with a panoramic view of wind turbines in an arid landscape, then moves to a pickup truck traveling through a wooded area, ending with a progressive distortion and blurring of the image, until a complete loss of definition.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while the snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Tilts to connect the snail’s path with the encroaching nature.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Tilting up from the snail’s rhythmic crawl to a rusted piece of wire fencing now acting as a trellis for wild ivy, the shallow focus underscores the beautiful decay of human boundaries.

OMC sidecar (001__osc_sc04_0150_comp_gen_a002_v001_sp_earth_porthole.1001.omc.json)

  • asset: osc_sc04_0150_comp_gen_a002_v001_sp_earth_porthole.1001.png
  • deliverable stem: osc_sc04_0150_comp_cmp_a003_v001_sp_earth_porthole.mov
  • era: Space
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1885895451
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:19:55.567Z
  • generation_seconds: 15
  • task_id: 019f1352-13f7-71c0-86ae-d8bd42fa3ed4
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 1cd2c016-d5fc-468a-80e0-c0307839989f
  • Kitsu entity id: 6383b942-74fe-40e9-959b-29ac7d2ed6f6 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0160

A smooth tracking shot from a moving vehicle captures a weathered ‘Old 66 Life’ sign surrounded by encroaching nature, emphasizing the serene reclamation of the historic road by the elements.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0160_comp_gen_a002_v001_hd_orbital_sky_tilt.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B03
  • beat_summary: Nature reclaims the asphalt below while the night sky above fills with the slow arcs of human satellites.
  • beat_shift: human abandonment -> natural reclamation
  • tempo: rapid
  • transition: Cut on Action

Camera

  • shot_size: Medium Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Tracking Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0326
  • archive_tc_in: 00:29:07:13
  • archive_tc_out: 1747.52

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Wide Shot; Eye Level; Tracking Shot; Rule of Thirds; Point-of-View Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a melancholic look, focusing on the medium wide tracking shot of the weathered ‘Old 66 Life’ sign surrounded by nature. Retime to 1.41 seconds. No generation.
  • reference_summary: The sequence begins with a shot from a car window following a silver caravan on a rural road, then moves to a road sign with the text ‘Old 66 Life’, maintained for subsequent scenes with slight background variations, suggesting a journey along the historic route-66.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Extreme wide shot to show the extent of the reclamation.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: A static, low-angle extreme wide shot places the tiny snail against the vast, empty sky and the crumbling horizon, emphasizing her enduring presence in a landscape that has forgotten humanity.

OMC sidecar (001__osc_sc04_0160_comp_gen_a002_v001_hd_orbital_sky_tilt.1001.omc.json)

  • asset: osc_sc04_0160_comp_gen_a002_v001_hd_orbital_sky_tilt.1001.png
  • deliverable stem: osc_sc04_0160_comp_cmp_a003_v001_hd_orbital_sky_tilt.mov
  • era: HD 2000s+
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1510110752
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:21:02.774Z
  • generation_seconds: 17
  • task_id: 019f1352-c7ea-7b92-b3a8-e2ce517ae543
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: bc751925-3cc1-451a-8fd1-513c6d5c7c34
  • Kitsu entity id: c36c23a1-b740-43c5-a2c9-a1b053ad639b (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0170

A static shot holds on a rusted ‘Old 66 Cafe’ sign as digital artifacts slowly distort the image, visually representing the fading memory of human presence as nature quietly takes over.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0170_comp_gen_a002_v001_hd_satellite_horizon.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B03
  • beat_summary: Nature reclaims the asphalt below while the night sky above fills with the slow arcs of human satellites.
  • beat_shift: human abandonment -> natural reclamation
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Medium Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0327
  • archive_tc_in: 00:29:12:14
  • archive_tc_out: 1752.56

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Medium Shot; Eye Level; Static Shot; Centered; Point-of-View Shot; Deep Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a melancholic, clinical look, focusing on the medium static shot of the rusted ‘Old 66 Cafe’ sign with digital artifacts. Retime to 1.41 seconds. No generation.
  • reference_summary: The sequence shows two distinct road signs related to route-66: first a cafe sign with the name ‘Old. 66 Cafe’, then an official ‘ROUTE US 66’ sign, which repeats in the third scene with an increase in digital visual disturbances.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Tracks her progress in a medium close-up, maintaining the frantic cut rate.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Tracking closely beside the snail, the camera captures the intricate textures of her shell contrasting with the rough, moss-covered chunks of broken asphalt she navigates.

OMC sidecar (001__osc_sc04_0170_comp_gen_a002_v001_hd_satellite_horizon.1001.omc.json)

  • asset: osc_sc04_0170_comp_gen_a002_v001_hd_satellite_horizon.1001.png
  • deliverable stem: osc_sc04_0170_comp_cmp_a003_v001_hd_satellite_horizon.mov
  • era: HD 2000s+
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 787567880
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:22:48.841Z
  • generation_seconds: 15
  • task_id: 019f1354-285b-7f93-8f82-08b3dd64bcbf
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 7b4a29fa-96f8-4913-919d-fd042f37511d
  • Kitsu entity id: aa6003fb-6658-4757-b821-914ba174fe70 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0180

A slow tilt up past a glowing, isolated route-66 neon sign reveals the vast, dark night sky above, bridging the fading earthly artifacts with the serene, starlit expanse.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0180_comp_gen_a002_v001_hd_snail_orbital_track.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B03
  • beat_summary: Nature reclaims the asphalt below while the night sky above fills with the slow arcs of human satellites.
  • beat_shift: human abandonment -> natural reclamation
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Wide Shot
  • camera_angle: Low Angle
  • camera_movement: Tilt
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0437
  • archive_tc_in: 00:40:24:17
  • archive_tc_out: 2424.68

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Wide Shot; Low Angle; Tilt; Rule of Thirds; Point-of-View Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip F157301_DREAMS_ROADS___STREET_OF_AMERICA to the cut. Regrade to a melancholic night look, focusing on the wide low-angle tilt up past the glowing Route 66 neon sign to the dark sky. Retime to 1.41 seconds. No generation.
  • reference_summary: The sequence shows a progression of neon signs against a dark background, starting with the letters ‘EL’, moving to a stylized figure with an arrow and ending with the iconic ‘ROUTE 66’ lettering, all characterized by strong color contrast and a lively night atmosphere.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B02
  • beat_summary: Nature slowly reclaims the cracked asphalt while snail continues her solitary trek.
  • beat_shift: Human dominance -> Natural reclamation
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pulls out to reveal her isolation before the abrupt location shift.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Pulling out slowly from a macro view of the snail’s face, the focus racks to reveal the vast, overgrown expanse of the dead highway stretching out before her, setting up the transition to the cosmos.

OMC sidecar (001__osc_sc04_0180_comp_gen_a002_v001_hd_snail_orbital_track.1001.omc.json)

  • asset: osc_sc04_0180_comp_gen_a002_v001_hd_snail_orbital_track.1001.png
  • deliverable stem: osc_sc04_0180_comp_cmp_a003_v001_hd_snail_orbital_track.mov
  • era: HD 2000s+
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1054525331
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:22:54.371Z
  • generation_seconds: 21
  • task_id: 019f1354-7cbd-73a3-b2e5-b1e31267637f
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 86cb2e69-8025-4ad5-839a-a526e796a135
  • Kitsu entity id: 3f463cfe-317f-46d8-bcee-2388626895ad (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0190

A wide panning shot follows a solitary yellow pickup truck driving through the vast, empty desert, its small presence dwarfed by the overwhelming scale of the natural landscape reclaiming the route.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0190_comp_gen_a002_v001_hd_snail_orbital_cu.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B03
  • beat_summary: Nature reclaims the asphalt below while the night sky above fills with the slow arcs of human satellites.
  • beat_shift: human abandonment -> natural reclamation
  • tempo: rapid
  • transition: Cut on Action

Camera

  • shot_size: Extreme Wide Shot
  • camera_angle: High Angle
  • camera_movement: Pan
  • lens: Ultra Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre#0016
  • archive_tc_in: 00:00:44:08
  • archive_tc_out: 44.32

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Extreme Wide Shot; High Angle; Pan; Rule of Thirds; Single Shot; Deep Depth of Field; Ultra Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip F719859_Kilimangiaro_Route_66_La_strada_madre to the cut. Regrade to a melancholic look, focusing on the extreme wide high-angle pan following the solitary yellow pickup truck in the desert. Retime to 1.41 seconds. No generation.
  • reference_summary: A sequence of three shots showing a yellow pickup truck with ‘Route US 66’ decorations moving on a road in a desert environment. The camera follows the vehicle from different angles: front, side and rear, highlighting the driver and the landscape context.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B03
  • beat_summary: The perspective abruptly shifts to the sterile, quiet interior of a lunar orbiting space capsule.
  • beat_shift: Earthly decay -> Cold technological triumph
  • kind: archive
  • tempo: rapid
  • duration_s: 0.65
  • editorial_note: Jarringly transports the viewer from the earthy road to the sterile lunar orbit.
  • archive_segment_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1#0103
  • source_archive_id: T12239_123_SPECIALE_TG1_2012__2013_Storia_dell__uomo_sulla_Luna_1
  • archive_in: 923.76
  • archive_out: 942.8
  • archive_tc_in: 00:15:23:19
  • archive_tc_out: 00:15:42:20
  • origin_code: arc
  • notes: Strong archive match (score 1.0); relations: conceptual, narrative
  • description: A stark, static wide shot of a lunar module floating silently in the void near the Moon, abruptly cutting from the earthly decay to the cold, clinical triumph of human spaceflight.
  • archive_summary_en: The sequence shows a space journey that begins with a lunar module near the Moon, continues with a panoramic view of the lunar surface and the Earth, and ends with the recovery of the space module at sea by a Navy aircraft.

OMC sidecar (001__osc_sc04_0190_comp_gen_a002_v001_hd_snail_orbital_cu.1001.omc.json)

  • asset: osc_sc04_0190_comp_gen_a002_v001_hd_snail_orbital_cu.1001.png
  • deliverable stem: osc_sc04_0190_comp_cmp_a003_v001_hd_snail_orbital_cu.mov
  • era: HD 2000s+
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 629510251
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:23:01.675Z
  • generation_seconds: 16
  • task_id: 019f1354-ea84-7fb0-b4b3-91d15b10c024
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: 373adc07-7418-4662-b5be-21ecabce8509
  • Kitsu entity id: 4ab2b00c-e282-4228-a258-57f4c1aee76a (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0200

A static close-up of a decorative route-66 plaque serves as a melancholic marker of the past, its fixed map of ‘America’s Main Street’ standing in stark contrast to the dynamic, overgrown reality of the present.

Preview

Latest Kitsu preview (revision 1, source: osc_sc04_0200_comp_gen_a002_v001_hd_highway_crane_climax.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B03
  • beat_summary: Nature reclaims the asphalt below while the night sky above fills with the slow arcs of human satellites.
  • beat_shift: human abandonment -> natural reclamation
  • tempo: rapid
  • transition: Match Cut

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: Roadbook#0054
  • archive_tc_in: 00:04:11:10
  • archive_tc_out: 251.4

Continuity

  • big framing jump (Extreme Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: Roadbook
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Close-Up Shot; Eye Level; Static Shot; Centered; Point-of-View Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Conform and grade archive clip Roadbook to the cut. Regrade to a melancholic, clinical look, focusing on the close-up static shot of the decorative Route 66 plaque. Retime to 1.41 seconds. No generation.
  • reference_summary: A static sequence showing a black decorative plaque with a map of the United States and Route 66 road signs for different states, with the text ‘America’s Main Street’ in the center. The cities of Chicago and Los Angeles are indicated. In the lower right there is a round object with the word …
  • brief notes: big framing jump (Extreme Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B03
  • beat_summary: The perspective abruptly shifts to the sterile, quiet interior of a lunar orbiting space capsule.
  • beat_shift: Earthly decay -> Cold technological triumph
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pushes in on the capsule interior to explore the new, alien environment.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Pushing in on the sterile, illuminated instrument panels of a space capsule interior, the high angle and shallow focus emphasize the detached, technological isolation of the orbiting craft.

OMC sidecar (001__osc_sc04_0200_comp_gen_a002_v001_hd_highway_crane_climax.1001.omc.json)

  • asset: osc_sc04_0200_comp_gen_a002_v001_hd_highway_crane_climax.1001.png
  • deliverable stem: osc_sc04_0200_comp_cmp_a003_v001_hd_highway_crane_climax.mov
  • era: HD 2000s+
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 408300037
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:24:27.509Z
  • generation_seconds: 13
  • task_id: 019f1355-ccd8-7d01-a9d7-36a5314495a5
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu preview file id: dd887973-c839-4224-8256-56c5fa1872a9
  • Kitsu entity id: 92bc8cc5-db0b-4a6a-93ed-1fbb0fe4524b (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0210

A historical shot of a bustling, formal street scene slowly degrades into heavy visual artifacts, the dissolving image poetically mirroring the physical decay and natural reclamation of the human world.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: archive
  • reference_archive: arc
  • beat: SCENE-04-B03
  • beat_summary: Nature reclaims the asphalt below while the night sky above fills with the slow arcs of human satellites.
  • beat_shift: human abandonment -> natural reclamation
  • tempo: rapid
  • transition: Jump Cut

Camera

  • shot_size: Wide Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 35
  • nb_frames: 35

Archive source

  • archive_id: C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO#0121
  • archive_tc_in: 00:17:15:15
  • archive_tc_out: 1035.6

Continuity

  • big framing jump (Close-Up Shot -> Wide Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: archive
  • method: Conform archive - no generation: use the cited real clip; regrade and retime it to the cut.
  • duration_s: 1.41
  • reference_archive_id: C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Wide Shot; Eye Level; Static Shot; Rule of Thirds; Group Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Conform and grade archive clip C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO to the cut. Regrade to a melancholic look, focusing on the wide static shot of the historical street scene degrading into visual artifacts. Retime to 1.41 seconds. No generation.
  • reference_summary: A sequence of a crowded street in a historical context, with people in formal clothes and vintage cars. The image quality degrades progressively, with visual artifacts increasing towards the end, making the right part of the image almost illegible.
  • brief notes: big framing jump (Close-Up Shot -> Wide Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B03
  • beat_summary: The perspective abruptly shifts to the sterile, quiet interior of a lunar orbiting space capsule.
  • beat_shift: Earthly decay -> Cold technological triumph
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Pans across the instruments, keeping the rapid montage pace alive.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Panning slowly across the cold, metallic interior of the capsule, the deep focus captures the stark, high-contrast lighting that mimics the harsh sunlight of the desolate lunar surface outside.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: Archival footage retrieved through ARC. See archive-footage-reuse.
  • Kitsu entity id: f2682c54-4c37-4769-a722-3d2f8ec8abd4 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc04_0220

The camera slowly cranes up from snail resting on a patch of weeds breaking through the asphalt, revealing a vast night sky where the slow, glowing arcs of satellites trace the new paths of human progress far above the forgotten road.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc04
  • status: running
  • canceled: yes
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-04-B03
  • beat_summary: Nature reclaims the asphalt below while the night sky above fills with the slow arcs of human satellites.
  • beat_shift: human abandonment -> natural reclamation
  • tempo: rapid
  • transition: Cut

Camera

  • shot_size: Extreme Wide Shot
  • camera_angle: High Angle
  • camera_movement: Crane Shot
  • lens: Ultra Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 70
  • nb_frames: 70

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. ROUTE 66, VARIOUS - MONTAGE / INT. SPACE CAPSULE - LUNAR ORBIT
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 2.81
  • reference_archive_id: C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO
  • style: Era/look: 1970s to Present. mood: Melancholic, Clinical, Accelerated. time of day: montage / int. space capsule - lunar orbit.
  • camera: Extreme Wide Shot; High Angle; Crane Shot; Rule of Thirds; Single Shot; Deep Depth of Field; Ultra Wide Angle Lens; Frame Rate
  • prompt: Extreme wide high-angle crane shot lifting up from a snail resting on weeds breaking through cracked asphalt. The deep depth of field reveals a vast, melancholic night sky where glowing arcs of satellites trace paths above the forgotten road. Ultra Wide Angle Lens, Rule of Thirds, 1970s-present era.
  • negative_prompt: shallow depth of field, static camera, low angle, bright daylight, crowded street on screen, visual artifacts on screen
  • reference_summary: A sequence of a crowded street in a historical context, with people in formal clothes and vintage cars. The image quality degrades progressively, with visual artifacts increasing towards the end, making the right part of the image almost illegible.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-04-B03
  • beat_summary: The perspective abruptly shifts to the sterile, quiet interior of a lunar orbiting space capsule.
  • beat_shift: Earthly decay -> Cold technological triumph
  • kind: synthetic
  • tempo: rapid
  • duration_s: 1.3
  • editorial_note: Extreme wide to emphasize the isolation of the space capsule.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: A static, extreme wide view from inside the capsule looks out through the window at the distant Earth, framing the sterile interior against the vast, silent panorama of space.

Notes

  • Part of sequence sc04 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu entity id: 9e45e14c-6074-4e18-b969-007d43e250d8 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc05 - 05 The End

Sequence sc05 - The End

Preview

First available Kitsu preview for this sequence (from shot sc05_0010). 5 of 6 shots in this sequence have Kitsu previews.

Facts

Shots

  • sc05_0010: A sweeping drone shot follows snail as she finally crosses the threshold onto the wooden planks of the Santa Monica Pier, the iconic…
  • sc05_0020: The camera pushes in slowly on the snail as she halts at the base of the wooden signpost, her tiny form dwarfed by the monumental ‘End’…
  • sc05_0030: An over-the-shoulder shot from behind the snail as she turns her gaze back toward the bustling pier and the long road behind her, the soft…
  • sc05_0040: The camera slowly arcs around the snail as she pivots away from the land, revealing the endless, sparkling expanse of the Pacific Ocean…
  • sc05_0050: A tight, static macro shot holds on the snail’s face as she finally rests her weary body against the wood, her relentless drive dissolving…
  • sc05_0060: The camera slowly pulls out and cranes up high above the pier, reducing the motionless snail to a tiny speck against the majestic,…

sc05_0010

A sweeping drone shot follows snail as she finally crosses the threshold onto the wooden planks of the Santa Monica Pier, the iconic ‘route-66 - End’ sign looming triumphantly against the bright California sky.

Preview

Latest Kitsu preview (revision 1, source: osc_sc05_0010_comp_gen_a002_v001_gh_end_sign_arrival.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc05
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-05-B01
  • beat_summary: The snail arrives at the ‘Route 66 - End’ sign at the Santa Monica Pier.
  • beat_shift: endless journey -> sudden finality
  • tempo: measured
  • transition: Cross Dissolve

Camera

  • shot_size: Wide Shot
  • camera_angle: High Angle
  • camera_movement: Drone Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 167
  • nb_frames: 167

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. SANTA MONICA PIER - LATE AFTERNOON
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 6.67
  • reference_archive_id: Roadbook
  • style: Era/look: Present. mood: Peaceful, Triumphant, Vast. time of day: late afternoon.
  • camera: Wide Shot; High Angle; Drone Shot; Leading Lines; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Wide high-angle drone shot sweeping over a snail as it crosses onto the wooden planks of the Santa Monica Pier. In the deep depth of field, the iconic ‘Route 66 - End’ sign looms triumphantly against a bright, peaceful late afternoon California sky. Wide Angle Lens, leading lines, present-day aesthetic.
  • negative_prompt: shallow depth of field, static camera, low angle, desert landscape, informative graphics on screen, dark gloomy lighting
  • reference_summary: Sequence of desert landscapes in Arizona, with the progressive addition of informative graphics indicating locations and distances, culminating with a wider view that includes a reference to Los Angeles and Santa Monica Pier in California.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-05-B01
  • beat_summary: The snail arrives at the ‘Route 66 - End’ sign on the Santa Monica Pier.
  • beat_shift: Endless journey -> Sudden arrival
  • kind: composite
  • tempo: slow
  • duration_s: 8.57
  • editorial_note: Slows the pace dramatically, blending into the present day as she reaches her destination.
  • archive_segment_id: Roadbook#0334
  • source_archive_id: Roadbook
  • archive_in: 1631.32
  • archive_out: 1640.96
  • archive_tc_in: 00:27:11:08
  • archive_tc_out: 00:27:20:24
  • origin_code: cmp
  • notes: Augment archive Roadbook with AI/VFX (score 0.5706)
  • description: Composited onto the wooden planks of the pier as the archival view widens to reveal the Santa Monica reference, the camera pushes in slowly on the snail as she finally reaches the iconic ‘Route 66 - End’ sign, her long journey culminating in this serene arrival.
  • archive_summary_en: Sequence of desert landscapes in Arizona, with the progressive addition of informative graphics indicating locations and distances, culminating with a wider view that includes a reference to Los Angeles and Santa Monica Pier in California.

OMC sidecar (001__osc_sc05_0010_comp_gen_a002_v001_gh_end_sign_arrival.1001.omc.json)

  • asset: osc_sc05_0010_comp_gen_a002_v001_gh_end_sign_arrival.1001.png
  • deliverable stem: osc_sc05_0010_comp_cmp_a003_v001_gh_end_sign_arrival.mov
  • era: era-locked-look
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1993909876
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:28:51.020Z
  • generation_seconds: 17
  • task_id: 019f1359-ecd6-7253-9715-0d0e218284f1
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc05 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 33656e6e-f0a5-4769-80cb-74dfcf27efa9
  • Kitsu entity id: 19434cbd-84a9-478f-86dd-4120b5bac15c (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:53.

sc05_0020

The camera pushes in slowly on snail as she halts at the base of the wooden signpost, her tiny form dwarfed by the monumental ‘End’ placard, capturing the sudden, overwhelming finality of her decades-long journey.

Preview

Latest Kitsu preview (revision 1, source: osc_sc05_0020_comp_gen_a002_v001_gh_end_sign_rest.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc05
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-05-B01
  • beat_summary: The snail arrives at the ‘route-66 - End’ sign at the Santa Monica Pier.
  • beat_shift: endless journey -> sudden finality
  • tempo: measured
  • transition: Eyeline Match

Camera

  • shot_size: Close-Up Shot
  • camera_angle: Low Angle
  • camera_movement: Push In
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 167
  • nb_frames: 167

Continuity

  • big framing jump (Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. SANTA MONICA PIER - LATE AFTERNOON
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 6.67
  • reference_archive_id: F719859_Kilimangiaro_Route_66_La_strada_madre
  • style: Era/look: Present. mood: Peaceful, Triumphant, Vast. time of day: late afternoon.
  • camera: Close-Up Shot; Low Angle; Push In; Centered; Single Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Close-up low-angle shot pushing in slowly on a snail halted at the base of a wooden signpost. The shallow depth of field blurs the monumental ‘End’ placard above, capturing a peaceful, triumphant finality in the late afternoon light. Centered composition, Medium Lens, present-day textures.
  • negative_prompt: deep depth of field, static camera, high angle, dark green automobile on screen, gloomy lighting, wide angle
  • reference_summary: A sequence of shots progressively showing the rear of a dark green automobile, with particular emphasis on the ‘ROUTE 66’ badge, the chrome strip and design details, including a taillight and a yellow sticker. The camera moves laterally, revealing more elements of the vehicle in each scene.
  • brief notes: big framing jump (Wide Shot -> Close-Up Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-05-B02
  • beat_summary: She looks back at the invisible, massive distance she has just conquered.
  • beat_shift: Forward momentum -> Retrospective awe
  • kind: synthetic
  • tempo: slow
  • duration_s: 12.86
  • editorial_note: Cuts to a medium close-up as she turns, dropping the pace to emphasize her reflection.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: Panning slowly from behind the snail’s weathered shell, the shallow focus looks back over her shoulder toward the blurred, golden-hour horizon, emphasizing the monumental, invisible distance she has just conquered.

OMC sidecar (001__osc_sc05_0020_comp_gen_a002_v001_gh_end_sign_rest.1001.omc.json)

  • asset: osc_sc05_0020_comp_gen_a002_v001_gh_end_sign_rest.1001.png
  • deliverable stem: osc_sc05_0020_comp_cmp_a003_v001_gh_end_sign_rest.mov
  • era: era-locked-look
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 2122217551
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:30:35.338Z
  • generation_seconds: 14
  • task_id: 019f135b-6ee1-7fd0-ade0-1c95aa97493c
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc05 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 5d650163-92d3-473f-9494-e365206c1a84
  • Kitsu entity id: 3aa56ff1-53a0-4ea6-942a-001c08f43db9 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:54.

sc05_0030

An over-the-shoulder shot from behind snail as she turns her gaze back toward the bustling pier and the long road behind her, the soft focus blurring the human world as she reflects on the invisible miles conquered.

Preview

Latest Kitsu preview (revision 1, source: osc_sc05_0030_comp_gen_a002_v001_gh_look_back.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc05
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-05-B02
  • beat_summary: She looks back at the invisible miles she has conquered, then turns to face the vast Pacific Ocean.
  • beat_shift: accomplishment -> existential peace
  • tempo: measured
  • transition: Cut

Camera

  • shot_size: Medium Close-Up Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 167
  • nb_frames: 167

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. SANTA MONICA PIER - LATE AFTERNOON
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 6.67
  • reference_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • style: Era/look: Present. mood: Peaceful, Triumphant, Vast. time of day: late afternoon.
  • camera: Medium Close-Up Shot; Eye Level; Static Shot; Rule of Thirds; Over-the-Shoulder Shot; Soft Focus; Medium Lens; Frame Rate
  • prompt: Medium close-up over-the-shoulder static shot from behind a snail at eye level. The snail turns its gaze back toward a bustling pier and the long road behind. The background is in soft focus, blurring the present-day human world into a peaceful, triumphant late afternoon glow. Rule of Thirds, Medium Lens.
  • negative_prompt: deep depth of field, moving camera, sharp background, interior store on screen, gloomy lighting, wide angle
  • reference_summary: A sequence of interactions in a store, with John Smith behind the counter assisting Lisa Brown during a purchase, while Mary Jones is initially present but no longer in subsequent scenes. The environment is a store with displayed items, natural lighting and a ‘route-66’ hat hanging from the ceiling.

Shot list (001__shot_list_v0.2.0_comment_v1_r06_revised.json (AGT-021 revised))

  • beat_id: SCENE-05-B03
  • beat_summary: She turns to face the vast Pacific Ocean and stops moving for the first time.
  • beat_shift: Restless drive -> Absolute stillness
  • kind: composite
  • tempo: slow
  • duration_s: 8.57
  • editorial_note: Fades the sequence to black, providing absolute finality to her long journey.
  • archive_segment_id: T08340_222_TG2_SI_VIAGGIARE_2008__2009_Viaggio_lungo_la_Route_66#0088
  • source_archive_id: T08340_222_TG2_SI_VIAGGIARE_2008__2009_Viaggio_lungo_la_Route_66
  • archive_in: 197.32
  • archive_out: 201.48
  • archive_tc_in: 00:03:17:08
  • archive_tc_out: 00:03:21:12
  • origin_code: cmp
  • notes: Augment archive T08340_222_TG2_SI_VIAGGIARE_2008__2009_Viaggio_lungo_la_Route_66 with AI/VFX (score 0.5133)
  • description: Fused onto the railing of the bustling pier against the vast backdrop of the Pacific Ocean, the camera holds perfectly static in an extreme wide shot as the snail finally ceases all movement, finding absolute peace at the edge of the continent.
  • archive_summary_en: The sequence begins with a close-up of a Santa Monica pier sign, then moves to a view along the pier with people and a moving motorcycle, showing the harbor and the sea in the background. The transition is fluid, with a slight shift in perspective to the right in the third scene.

OMC sidecar (001__osc_sc05_0030_comp_gen_a002_v001_gh_look_back.1001.omc.json)

  • asset: osc_sc05_0030_comp_gen_a002_v001_gh_look_back.1001.png
  • deliverable stem: osc_sc05_0030_comp_cmp_a003_v001_gh_look_back.mov
  • era: era-locked-look
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 771687640
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:32:05.261Z
  • generation_seconds: 13
  • task_id: 019f135c-e7bc-7e30-955b-9849c6928595
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc05 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 5e434293-9c4e-4d97-bab3-b30119d1427a
  • Kitsu entity id: e4d5c103-6f79-439d-bc09-2d1fbd18c649 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:54.

sc05_0040

The camera slowly arcs around snail as she pivots away from the land, revealing the endless, sparkling expanse of the Pacific Ocean stretching out before her, marking her transition to existential peace.

Preview

Latest Kitsu preview (revision 1, source: osc_sc05_0040_comp_gen_a002_v001_gh_road_traversed.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc05
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-05-B02
  • beat_summary: She looks back at the invisible miles she has conquered, then turns to face the vast Pacific Ocean.
  • beat_shift: accomplishment -> existential peace
  • tempo: measured
  • transition: Cut on Action

Camera

  • shot_size: Medium Shot
  • camera_angle: Eye Level
  • camera_movement: Arc Shot
  • lens: Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 167
  • nb_frames: 167

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. SANTA MONICA PIER - LATE AFTERNOON
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 6.67
  • reference_archive_id: Roadbook
  • style: Era/look: Present. mood: Peaceful, Triumphant, Vast. time of day: late afternoon.
  • camera: Medium Shot; Eye Level; Arc Shot; Dynamic Symmetry; Single Shot; Deep Depth of Field; Wide Angle Lens; Frame Rate
  • prompt: Medium eye-level shot slowly arcing around a snail as it pivots away from the land. The deep depth of field reveals the endless, sparkling expanse of the Pacific Ocean stretching out before it. The scene is bathed in peaceful, triumphant late afternoon light. Dynamic symmetry, Wide Angle Lens, present-day.
  • negative_prompt: shallow depth of field, static camera, desert landscape, informative graphics on screen, gloomy lighting, close-up
  • reference_summary: Sequence of desert landscapes in Arizona, with the progressive addition of informative graphics indicating locations and distances, culminating with a wider view that includes a reference to Los Angeles and Santa Monica Pier in California.

Shot list (001__shot_list.json (AGT-020))

  • beat_id: SCENE-05-B02
  • beat_summary: She looks back at the invisible, conquered road behind her.
  • beat_shift: forward momentum -> retrospective awe
  • kind: composite
  • tempo: brisk
  • duration_s: 3.75
  • editorial_note: Pulls out slightly to reveal the invisible road behind her.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0002
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 14.96
  • archive_out: 31.68
  • origin_code: cmp
  • notes: Augment archive F157301_DREAMS_ROADS___STREET_OF_AMERICA with AI/VFX (score 0.5231)
  • description: Pulling out slowly from the snail perched on the stylized map graphic, the camera reveals the entire illustrated route she has traversed, visually cementing her pride and the monumental scale of her achievement.
  • archive_summary_en: The sequence begins with an abstract colored background with the ‘EMERSON COMPANY’ logo, then introduces the title ‘ROUTE 66 “The Mother Road”’ and the director’s name ‘Franco Gattafoni’ in a central box, concluding with a stylized representation of a map showing the route-66 road, keeping the …

OMC sidecar (001__osc_sc05_0040_comp_gen_a002_v001_gh_road_traversed.1001.omc.json)

  • asset: osc_sc05_0040_comp_gen_a002_v001_gh_road_traversed.1001.png
  • deliverable stem: osc_sc05_0040_comp_cmp_a003_v001_gh_road_traversed.mov
  • era: era-locked-look
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 232967413
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:33:30.579Z
  • generation_seconds: 12
  • task_id: 019f135e-3a2a-7613-b3f0-e036e3c961c8
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc05 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 25b797e9-7c41-481a-967d-533678085395
  • Kitsu entity id: 873ffcb7-093e-4efe-8611-b89c9ba2c7b0 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:54.

sc05_0050

A tight, static macro shot holds on snail’s face as she finally rests her weary body against the wood, her relentless drive dissolving into ultimate stillness as the era-locked-look light washes over her.

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc05
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-05-B03
  • beat_summary: For the first time, the snail stops moving entirely as the camera pulls back to reveal the ocean and the sky.
  • beat_shift: relentless drive -> ultimate stillness
  • tempo: measured
  • transition: Match Cut

Camera

  • shot_size: Macro Shot
  • camera_angle: Eye Level
  • camera_movement: Static Shot
  • lens: Medium Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 167
  • nb_frames: 167

Continuity

  • big framing jump (Medium Shot -> Macro Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. SANTA MONICA PIER - LATE AFTERNOON
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 6.67
  • reference_archive_id: C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO
  • style: Era/look: Present. mood: Peaceful, Triumphant, Vast. time of day: late afternoon.
  • camera: Macro Shot; Eye Level; Static Shot; Centered; Single Shot; Shallow Depth of Field; Medium Lens; Frame Rate
  • prompt: Macro static shot at eye level holding tightly on a snail’s face as it rests against weathered wood. The shallow depth of field isolates the snail, bathed in peaceful, triumphant golden hour late afternoon light, dissolving into ultimate stillness. Centered composition, Medium Lens, present-day textures.
  • negative_prompt: deep depth of field, moving camera, wide angle, human faces on screen, gloomy lighting, modern sleek surfaces
  • reference_summary: The sequence shows a woman identified as Marina, initially in a scene with a man, then increasingly in the foreground up to a close-up facial portrait, while the man progressively disappears from the frame. The atmosphere is serious and reflective, with a strong focus on the woman.
  • brief notes: big framing jump (Medium Shot -> Macro Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list.json (AGT-020))

  • beat_id: SCENE-05-B03
  • beat_summary: She turns to the vast ocean, breathes, and finally stops moving.
  • beat_shift: relentless drive -> profound stillness
  • kind: composite
  • tempo: brisk
  • duration_s: 3.75
  • editorial_note: Focuses intimately on her final breath as she stops moving.
  • archive_segment_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA#0055
  • source_archive_id: F157301_DREAMS_ROADS___STREET_OF_AMERICA
  • archive_in: 286.48
  • archive_out: 300.92
  • origin_code: cmp
  • notes: Augment archive F157301_DREAMS_ROADS___STREET_OF_AMERICA with AI/VFX (score 0.5231)
  • description: A static, extreme close-up captures the snail coming to a complete, profound halt against the cloudy blue background and the names of fellow dreamers, her stillness radiating absolute peace.
  • archive_summary_en: A sequence of three scenes with a blue background similar to a cloudy sky and a vertical gray box on the left. Initially, a dedication appears to those who have the courage and strength to follow their dreams. Subsequently, the names of four influential figures are added: John Steinbeck, Woody …

OMC sidecar (001__osc_sc05_0050_comp_gen_a002_v001_gh_ocean_stillness.1001.omc.json)

  • asset: osc_sc05_0050_comp_gen_a002_v001_gh_ocean_stillness.1001.png
  • deliverable stem: osc_sc05_0050_comp_cmp_a003_v001_gh_ocean_stillness.mov
  • era: Golden Hour
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 1294186795
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:35:10.952Z
  • generation_seconds: 12
  • task_id: 019f135f-be34-71e2-86cf-d6a4017a7a16
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc05 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 28227d62-d055-4a9b-ade0-f2441f90eb01
  • Kitsu entity id: 557102b3-4dd8-4b42-a0d5-79b120380090 (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:54.

sc05_0060

The camera slowly pulls out and cranes up high above the pier, reducing the motionless snail to a tiny speck against the majestic, boundless horizon of the Pacific Ocean and the setting sun, bringing the epic journey to a serene close.

Preview

Latest Kitsu preview (revision 1, source: osc_sc05_0060_comp_gen_a002_v001_gh_ocean_turn.1001).

Shot data (kitsu)

  • Kitsu: open in Kitsu
  • sequence: sc05
  • status: running
  • canceled: no
  • kind: synthetic
  • reference_archive: gen
  • beat: SCENE-05-B03
  • beat_summary: For the first time, snail stops moving entirely as the camera pulls back to reveal the ocean and the sky.
  • beat_shift: relentless drive -> ultimate stillness
  • tempo: measured
  • transition: Fade Out

Camera

  • shot_size: Extreme Wide Shot
  • camera_angle: High Angle
  • camera_movement: Pull Out
  • lens: Ultra Wide Angle Lens

Timing

  • fps: 25
  • frame_in: 1
  • frame_out: 167
  • nb_frames: 167

Continuity

  • big framing jump (Macro Shot -> Extreme Wide Shot) - add an intermediate size or motivate the cut

movielabs export (JSON)

Generation brief (001__generation_briefs_v0.3.0_r02.json)

  • scene_heading: EXT. SANTA MONICA PIER - LATE AFTERNOON
  • kind: synthetic
  • method: Generate (image/video-to-video) - archive-grounded: the cited clip is a visual reference for look and period detail, not reused on screen.
  • duration_s: 6.67
  • reference_archive_id: C10371_TECNOLOGIA__MOMENTI_DEL_PROGRESSO
  • style: Era/look: Present. mood: Peaceful, Triumphant, Vast. time of day: late afternoon.
  • camera: Extreme Wide Shot; High Angle; Pull Out; Rule of Thirds; Single Shot; Deep Depth of Field; Ultra Wide Angle Lens; Frame Rate
  • prompt: Extreme wide high-angle shot slowly pulling out and craning high above the Santa Monica pier. The motionless snail is reduced to a tiny speck against the majestic, deep depth of field horizon of the Pacific Ocean and the setting sun. Peaceful, triumphant late afternoon light. Ultra Wide Angle Lens, Rule of Thirds.
  • negative_prompt: shallow depth of field, static camera, close-up, human faces on screen, gloomy lighting, interior setting
  • reference_summary: The sequence shows a man identified as John Doe in three progressive shots, which pass from a scene set with objects in the background to an increasingly tight close-up, with a growing focus on the face and a reduction of the surrounding context. The light and composition change to emphasize the …
  • brief notes: big framing jump (Macro Shot -> Extreme Wide Shot) - add an intermediate size or motivate the cut

Shot list (001__shot_list.json (AGT-020))

  • beat_id: SCENE-05-B03
  • beat_summary: She turns to the vast ocean, breathes, and finally stops moving.
  • beat_shift: relentless drive -> profound stillness
  • kind: synthetic
  • tempo: measured
  • duration_s: 5.62
  • editorial_note: Arcs around her as she turns to face the vast ocean.
  • origin_code: gen
  • notes: No suitable archive match; generate synthetically per era lock.
  • description: The camera arcs slowly around the snail as she turns her gaze toward the shimmering Pacific horizon, the gentle movement emphasizing her transition from relentless drive to a state of profound, breathing stillness.

OMC sidecar (001__osc_sc05_0060_comp_gen_a002_v001_gh_ocean_turn.1001.omc.json)

  • asset: osc_sc05_0060_comp_gen_a002_v001_gh_ocean_turn.1001.png
  • deliverable stem: osc_sc05_0060_comp_cmp_a003_v001_gh_ocean_turn.mov
  • era: era-locked-look
  • model: fal-ai/nano-banana-2 (Gen-Image)
  • model_foundation: Google gemini (Nano Banana 2)
  • provider: fal
  • seed: 173476260
  • aspect_ratio: 16:9
  • resolution: 1K
  • generated_at: 2026-06-29T12:37:01.785Z
  • generation_seconds: 15
  • task_id: 019f1361-8663-7753-988b-14cf7454ad0a
  • governor agents: Narrative, Style, Temporal
  • reproducibility: Outputs from the gemini-family model (Nano Banana) are close but NOT bit-identical between runs, even with a fixed seed - expect minor variation when regenerating. Every generated image carries an invisible SynthID watermark marking it as AI-generated. Reference conditioning is text-mediated: …

Notes

  • Part of sequence sc05 in production one-small-crawl.
  • Origin: AI-generated / synthetic content (gen). See ai-agent-pipeline.
  • Kitsu preview file id: 221f7ba6-4873-48c4-aa3f-09cf60df3222
  • Kitsu entity id: b5fd9693-f434-4f19-a501-b24985b1addd (type 4b39ab64-70e8-4f53-aac4-e01f4878921a); created 2026-06-19T09:19:54.

Tools

  • Adobe After Effects: Adobe After Effects compositing/motion
  • Blender: open-source 3D creation suite
  • ChatGPT (OpenAI): ChatGPT (OpenAI) assistant
  • Edit Mind: Edit Mind editing assistant
  • Fal.ai: Fal.ai generative media API
  • Gemini (Google): Google Gemini (LLM/vision)
  • Google Cloud: Google Cloud platform
  • GraphDB: GraphDB knowledge-graph store
  • HOLLM: AI production assistant (HOLLI/mediadata); manages the Production Brain
  • Kitsu: open-source production tracker (CGWire); shot list + review backbone
  • LFM: LLM run via Unsloth for HOLLM
  • LoRA (Low-Rank Adaptation): LoRA (Low-Rank Adaptation) model fine-tuning
  • Maya: Autodesk Maya 3D animation
  • Moments Lab: Moments Lab media intelligence
  • Ornith: LLM run via Unsloth for HOLLM
  • Adobe Premiere Pro: Adobe Premiere Pro NLE
  • Qwen: Alibaba LLM run via Unsloth for HOLLM
  • RAI CRITS: RAI CRITS tooling
  • Tools - render_youtube_final.py (DaVinci Resolve scripting): Resolve render script for the final YouTube export
  • DaVinci Resolve: DaVinci Resolve edit/colour/delivery pipeline
  • Slack: Slack team communication
  • Suno: Suno AI music generation
  • Topaz (Gigapixel AI / Video AI): Topaz Gigapixel/Video AI upscaling + enhancement
  • Unreal Engine: Unreal Engine real-time renderer
  • Unsloth: open-source LLM inference tool (runs Ornith/LFM/Qwen for HOLLM)

Section cover (generated by the wiki book build script). See the Index for one-line summaries.

Adobe After Effects

Adobe’s digital visual effects, motion graphics, and compositing application. Part of the Adobe Creative Cloud. On One Small Crawl it appears in the compositing / motion-graphics discussion alongside Resolve and Blender.

  • Vendor: Adobe (Creative Cloud).
  • Use: motion graphics, VFX, compositing, title / graphics work, keyed composites.

On One Small Crawl

  • resolve – DaVinci Resolve for edit / colour / delivery
  • blender – 3D / VFX tool alongside Resolve
  • topaz – Topaz upscaling / enhancement

Blender

Free and open-source 3D computer graphics software suite for modeling, animation, rendering, compositing, motion tracking, VFX, 2D animation, and video editing. Originally an in-house tool at Dutch studio NeoGeo, publicly released in 1995; now developed by the Blender Foundation and community.

  • Author / origin: Ton Roosendaal; developed for NeoGeo (1987–1991 as “Traces”, then Blender).
  • Public release: 2 January 1994 (start); 1.00 January 1995; open-sourced / free online release 1 January 1998.
  • Stable release: 5.2 (14 July 2026).
  • Written in: C++, Python.
  • License: GPL-2.0-or-later.
  • Platforms: Windows, macOS, Linux, BSD, Haiku, IRIX.
  • Website: blender.org / projects.blender.org/blender.
  • Extensions: Blender Extensions repo introduced in 4.2 (add-ons + themes installable inside Blender).

Blender is used for animated films, VFX, art, 3D-printed models, motion graphics, VR, and game assets. It carries a robust Python API and is widely used in education (e.g., University of Michigan) and for synthetic image generation for computer-vision / AI training.

On One Small Crawl

ChatGPT (OpenAI)

OpenAI’s ChatGPT large-language-model assistant, used on this production as part of the metadata-analysis workflow and for shot-list / shot-blueprint ideation.

ChatGPT is a conversational LLM interface over OpenAI’s GPT model family (GPT-3.5, GPT-4, GPT-4o, o-series reasoning models). Used for drafting, brainstorming, metadata extraction, and light-shot-list prototyping.

On One Small Crawl

  • 2026-05-11-frames-call – the metadata-analysis workflow (Edit Mind JSON + ChatGPT) was scaled up (2026-05-11-frames-call).
  • Paired with Edit Mind (edit-mind) for metadata analysis; the ChatGPT piece feeds structured/semi-structured metadata out of the Edit Mind JSON.
  • edit-mind – Edit Mind, the metadata-analysis / JSON tool ChatGPT was paired with
  • gemini – Gemini shotgun / shot-blueprint tool (the later Google-based parallel)
  • ai-agent-pipeline – ARC + generation pipeline

Edit Mind

Metadata-analysis and transcription tool referenced in the FRAMES calls, used to structure editorial metadata (Edit Mind JSON) at scale.

On One Small Crawl

  • chatgpt – ChatGPT paired with Edit Mind JSON in the metadata-analysis workflow
  • lora – LoRA models for snail ages
  • rai-crits – RAI CRITS, chosen over Edit Mind on 2026-05-18-frames-call
  • moments-lab – Moments Lab, considered for content retrieval + semantic analysis

Edit Mind - editing assistant (no Wikipedia article; official site)

Fal.ai

Serverless AI inference platform / API gateway for running generative models – image generation, video generation, upscaling, and other model workloads – on demand, often as an alternative to running models locally.

On One Small Crawl, Fal was referenced once, on 2026-06-08-frames-call, alongside the Gemini shot list (gemini) and the Topaz enhancement path (topaz), in the shot-list / shot-blueprint discussion. Fal is the kind of inference/API layer that can host a LoRA-loaded diffusion model or an upscaling model on demand; whether it was adopted, tested, or merely considered is per-call.

On One Small Crawl

  • 2026-06-08-frames-call – Fal referenced on the call (2026-06-08-frames-call): the shot-list / shot-blueprint discussion mentions Fal alongside Gemini and Topaz. This is consistent with the production evaluating API-based inference (Fal) vs local inference (the local-laptop image-generation agent Cedric ran for a demo on 2026-06-22-frames-call) vs model-in-editor workflows.
  • gemini – Gemini shot-list / shot-blueprint tool
  • topaz – Topaz upscaling / enhancement (local tool path)
  • lora – LoRA character-consistency adapters that could run locally or via an inference API
  • ai-agent-pipeline – ARC + generation pipeline

fal.ai - generative media API (no Wikipedia article; official site)

Gemini (Google)

Google’s family of large language models / generative AI assistants, used on this production for shot-list and shot-blueprint generation.

Gemini is a family of multimodal LLMs from Google DeepMind, spanning the Gemini model family (Gemini 1.0, 1.5, 2.0 lines) with capabilities spanning text, image, audio, and video understanding. Used via the Google AI Studio / Gemini API and in Google Workspace integrations.

On One Small Crawl

  • topaz – Topaz upscaling / enhancement of generated frames
  • eddie-ai – Eddie AI, agentic AI editor proposed for rough cuts
  • ai-generated-frames – Topaz TIFFs + prob4_hyp1 clips
  • ai-assisted-post – Topaz / prob4_hyp1 enhancement evidence

Google Cloud

Google Cloud Platform (GCP) – the cloud infrastructure layer discussed on One Small Crawl for hosting, storage, compute, and possibly ML/inference.

GCP is Google’s cloud platform (Compute Engine, Cloud Storage, Vertex AI, BigQuery, Cloud Run, etc.). On One Small Crawl it came up repeatedly as the “cloud” in the “Google Cloud” references – likely the storage + compute / ML backbone for generated assets, frame storage, or inference hosting.

On One Small Crawl

The recurring mentions across four calls suggest GCP is the production’s cloud backbone – storage of generated frames / assets, possibly ML inference hosting (Vertex AI or a self-hosted model endpoint), and collaboration infrastructure with the remote team.

  • kitsu – production tracker (runs locally at localhost:5000; Docker environment)
  • gemini – Google’s Gemini LLM (a Google service, possibly consumed via Google Cloud / Google AI Studio)
  • ai-agent-pipeline – ARC + generation pipeline (generation compute / storage may land on Google Cloud)
  • ai-generated-frames – AI-generated frames asset (Topaz TIFFs + prob4_hyp1 clips)

GraphDB

Semantic knowledge-graph / RDF triplestore used as a source of truth on the One Small Crawl production, alongside Kitsu (kitsu), for structured metadata, ontologies, and entity relationships.

GraphDB is a semantic graph database / RDF triplestore (by Ontotext) used to store, query, and reason over RDF data – triples of subject–predicate–object – often backing ontologies, knowledge graphs, and linked-data applications. It is the “GraphDB” referenced on the production as part of the Kitsu + GraphDB source-of-truth architecture discussed on 2026-06-25-frames-call.

  • Type: RDF triplestore / semantic knowledge graph.
  • Role on One Small Crawl: structured metadata and ontology backbone – part of the “Kitsu + GraphDB source of truth” architecture from 2026-06-25-frames-call.
  • Related ontologies/tools in the wiki: omc-ontology (MovieLabs OMC ontology), the show-ontology x PeriodO crosswalk (show-ont-periodo-crosswalk), the asset-naming schema (naming-schema).

On One Small Crawl

  • 2026-06-25-frames-call – “Kitsu + GraphDB source of truth” surfaced on the call (2026-06-25-frames-call). Kitsu tracks the production side (shots, assets, tasks); GraphDB is the structured-knowledge / ontology layer that ties entities, metadata, and crosswalks together.
  • The wiki already carries the MovieLabs OMC ontology page (omc-ontology), the OSC Asset Naming Schema (naming-schema), the show-ontology x PeriodO crosswalk (show-ont-periodo-crosswalk), and related stubs – the GraphDB layer is the infrastructurebehind these structured-knowledge assets.
  • kitsu – production tracker (shot/asset/task side of the source of truth)
  • omc-ontology – MovieLabs OMC ontology
  • naming-schema – OSC Asset Naming Schema v1.0
  • show-ont-periodo-crosswalk – Show Ontology x PeriodO Crosswalk v001
  • rag – (if used) the RAG / retrieval-augmented side feeding from the graph

GraphDB - Ontotext RDF triplestore (no Wikipedia article; official site)

HOLLM

HOLLM is the AI production assistant developed by holli for the mediadata project at insa. It manages the One Small Crawl Production Brain and connects it to the different tools on the production.

  • Developer: holli (for the mediadata project at insa).
  • Type: AI production assistant / agentic orchestration layer.
  • Based on: Hermes Agent (by Nous Research).
  • Extends: using tools, skills and MCP from www.hectools.com.
  • Licence: open source.
  • On this production: manages the One Small Crawl Production Brain and connects it to the production’s tools.

What it does

  • Production Brain - HOLLM manages the One Small Crawl Production Brain, the central knowledge/state store for the production. It is the articulation between an LLM and a semantic wiki based on MovieLabs OMC (omc-ontology) and industry good practices (ves-good-practices).
  • Tool orchestration - connects the Production Brain to the different tools on the production (kitsu, resolve, graphdb, and the agent pipeline), so the brain stays in sync with what the tools do.
  • Agentic assistant - as an AI production assistant, it drives the agent-orchestrated workflow of one-small-crawl (see ai-agent-pipeline).

Hardware and inference (IBC Accelerator FRAMES demo)

For the IBC Accelerator FRAMES demo, HOLLM is powered by an HP Z2 mini workstation provided by hp, using an AMD Strix Halo processor (amd), running the unsloth inference tool for the ornith, lfm and qwen models.

  • Workstation: HP Z2 mini (provided by hp).
  • Processor: AMD Strix Halo (amd).
  • Inference: unsloth inference tool.
  • Models: ornith, lfm, qwen.

On One Small Crawl

  • HOLLM is the assistant layer that ties the Production Brain to the tool stack; it is the HOLLI-side counterpart to the agent-orchestrated pipeline described in the agent-production-bible.
  • It is developed under the mediadata project, a partnership between holli and insa for which cedric-lejeune is the tech manager (see cedric-lejeune).
  • The demo hardware (HP Z2 mini + AMD Strix Halo) is part of the AMD/HP kit supplied to the FRAMES proof-of-concept (see hp, amd).

HOLLM - HOLLI’s AI production assistant (no Wikipedia article; official site)

Kitsu

Open-source production tracker for animation, VFX, and game studios, developed by CGWire. Brings teams, assets, shots, reviews, and tasks into a single place so distributed teams can collaborate and ship faster.

  • Developer: CGWire.
  • Type: open-source production management system (frontend + Zou API backend, distributed as a Docker image).
  • Website: kitsu.cloud.
  • On this production: running locally at localhost:5000 (project “One Small Crawl”).

What it tracks

  • Shots – individual camera angles/scenes organized into sequences; each shot carries technical specs (frames, resolution, FPS), frame in/out, max retakes, and per-task status (Layout, Animation, Lighting, Compositing, …).
  • Assets – characters, props, environments, with custom asset types and task types.
  • Tasks & status – per-shot, per-department task status with colour-coded progress, artist avatars, time spent / estimation.
  • Reviews – upload previews, supervisor review, comments, H264 playback.
  • Import/export – CSV export of shot lists (sequence+shot names, specs, metadata, statuses, assignations, time tracking); EDL timecode import for frame in/out.

On One Small Crawl

Kitsu is the authoritative shot list and review backbone for the production:

  • Set up – Kitsu environment set up by cedric-lejeune on 2026-06-25-frames-call: manages sequences, shots, reviews, H264 playback, comments; Kitsu API import/export tested with chris-vienneau; Kitsu Slack connector investigated; NemoClaw integration checked.
  • Demoed – Kitsu workflow demoed (Docker) on 2026-06-01-frames-call, then presented again on 2026-06-08-frames-call (2026-06-08-frames-call, 2026-06-01-frames-call).
  • Shot list – the 5 sequences (sc01–sc05) and their 50+ shots on this production were pulled from the local Kitsu instance and turned into markdown pages (2026-09-02, see log). Shot pages carry full Kitsu data; 18 of the 50 shots are canceled in Kitsu (sc03_0060–0100, sc04_0100–0220).
  • Team – 11–12 Kitsu accounts on the production (team); cedric-lejeune is the instance admin (“Super Admin” role label on Kitsu).
  • Milestones – Kitsu milestones for the sizzle deadline discussed on 2026-08-17-frames-call.

Kitsu kind vs export kind

Notes on conflicts: Kitsu kind vs export kind differ on some shots; export records are kept verbatim alongside Kitsu data (see log). This is tracked per shot page in entities/production/shots/.

  • render-youtube-final-py – Resolve render script that exports the final timeline
  • resolve – DaVinci Resolve edit/colour/delivery pipeline
  • topaz – Topaz upscaling/enhancement feeding generated/archival frames
  • entities/production/shots/ (see the sequence pages under sc01-sc05) – shot pages pulled from local Kitsu
  • cedric-lejeune – manages Kitsu setup and Resolve post/colour
  • team – 11 Kitsu accounts on the production

Kitsu - CGWire open-source production tracker (no Wikipedia article; official site)

LFM

LFM is an LLM run on One Small Crawl via unsloth for hollm.

  • Type: language model.
  • On this production: one of the models hollm runs via unsloth on the HP Z2 mini (hp) with the AMD Strix Halo processor (amd).

LFM - Liquid AI foundation models (no Wikipedia article; official site)

LoRA (Low-Rank Adaptation)

Low-Rank Adaptation – a lightweight method for fine-tuning large pretrained models (especially diffusion image models and LLMs) by injecting small trainable low-rank matrices into frozen weights, so you can specialize a model (e.g., to a character) with a tiny fraction of the parameters and compute of a full fine-tune.

LoRA is widely used to train consistent character / style adapters on image-generation models (Stable Diffusion and its forks). A LoRA trained on a character produces a reusable adapter that can be loaded into a base model at inference time to get that character’s look across poses, lighting, and eras while keeping the base model frozen.

On One Small Crawl

LoRA work on the snail is a throughline of the AI generation pipeline (ai-agent-pipeline) – the character-consistency backbone that lets the snail appear across eras while still reading as the same character.

Maya

Autodesk Maya – professional 3D computer graphics application for 3D modeling, animation, rigging, VFX, and rendering. Industry-standard in animation and VFX pipelines. On One Small Crawl it appears across several FRAMES calls as part of the 3D/animation tool conversation.

  • Vendor: Autodesk.
  • Use: 3D modeling, character animation, rigging, VFX, rendering, pipeline scripting (Python / MEL).
  • Position in the production: potential 3D / character / animation tool alongside Blender (blender); Maya and Blender are the two 3D tools that surface in the calls.

On One Small Crawl

Maya is one of the 3D tools the production has in view – alongside Blender (blender) and the Unreal renders (unreal-render-pipeline). For the snail character and the Route 66-era 3D work, the production is weighing Blender (open-source, already in use for concept/blocking) vs Maya (industry-standard rigging/animation).

  • blender – Blender, the open-source 3D tool in use on the production
  • resolve – DaVinci Resolve for edit / colour / delivery
  • unreal-render-pipeline – Unreal Engine renders (DesertSequence)
  • ai-agent-pipeline – ARC + generation pipeline (LoRA snail, generation)

Moments Lab

Content retrieval and semantic-analysis platform for video / audio / image archives. On One Small Crawl it was considered as the content-retrieval + semantic-analysis tool for the RAI archive and other external archives (BBC, ITV, ITN, Route 66 museums).

On One Small Crawl

  • 2026-04-13-frames-call – Moments Lab considered for content retrieval + semantic analysis of the RAI archive and external archives (BBC, ITV, ITN, Route 66 museums) (2026-04-13-frames-call). Action item 4: reach out to the external archives; Moments Lab was the tool candidate for content retrieval + semantic analysis.
  • Later, on 2026-05-18-frames-call, RAI CRITS was chosen over Edit Mind for the metadata-analysis workflow; whether Moments Lab stayed in the picture for the archive-retrieval side is tracked per call.

Moments Lab - media intelligence (no Wikipedia article; official site)

Ornith

Ornith is an LLM run on One Small Crawl via unsloth for hollm.

  • Type: language model.
  • On this production: one of the models hollm runs via unsloth on the HP Z2 mini (hp) with the AMD Strix Halo processor (amd).
  • hollm - the assistant that runs this model
  • unsloth - inference tool
  • lfm, qwen - sibling models

Adobe Premiere Pro

Adobe’s professional nonlinear video editing application, part of the Adobe Creative Cloud. On One Small Crawl it was floated as an alternative editing route alongside DaVinci Resolve (resolve).

  • Vendor: Adobe (Creative Cloud).
  • Use: nonlinear editing, timeline-based assembly, multicam, basic colour, audio, titling.

On One Small Crawl

  • 2026-06-01-frames-call – the premiere route discussed on the Kitsu demo / Google Cloud call (2026-06-01-frames-call): “premiere route” came up as an option in the post / edit discussion. The decision land on Resolve (resolve) as the primary edit/colour/delivery tool on the production.
  • resolve – DaVinci Resolve, the primary edit / colour / delivery tool on the production
  • kitsu – production tracker feeding shot lists into the edit
  • after-effects – After Effects, in the Scene 3 animatic / compositing discussion

Qwen

Qwen is an LLM family (Alibaba) run on One Small Crawl via unsloth for hollm.

  • Type: language model (Alibaba Qwen family).
  • On this production: one of the models hollm runs via unsloth on the HP Z2 mini (hp) with the AMD Strix Halo processor (amd).

RAI CRITS

RAI archive research / content-analysis tool used on One Small Crawl. Chosen over Edit Mind (edit-mind) on 2026-05-18-frames-call as the metadata-analysis / content-retrieval tool for the RAI archive and the broader external-archive research workflow.

On One Small Crawl

  • 2026-05-18-frames-call – after LoRA models for snail ages were done, RAI CRITS was chosen over Edit Mind for the metadata-analysis workflow (2026-05-18-frames-call). From this point the production’s metadata / archive-research tooling pivots to RAI CRITS.
  • Implied broader role: content retrieval and semantic analysis of the RAI archive (the Italian national broadcaster archive) – the production’s primary European archive source – and possibly other external archives (the call earlier considered Moments Lab for the same purpose, moments-lab).
  • edit-mind – Edit Mind, the metadata-analysis tool RAI CRITS replaced
  • moments-lab – Moments Lab, earlier candidate for content retrieval + semantic analysis
  • artemis-ii-tg3-news – RAI news archive asset (Artemis II, Tg3)
  • 2026-04-13-frames-call – the call where archive access strategy was set (RAI material usable for research within IBC group; public/scientific use needs explicit authorization)
  • 2026-05-18-frames-call – the call where RAI CRITS was chosen over Edit Mind

RAI CRITS - Rai Research Center (no Wikipedia article; official site)

Tools - render_youtube_final.py (DaVinci Resolve Scripting)

Observed asset. A Python script that drives DaVinci Resolve’s API to export the current timeline with the “YouTube” preset.

What it does

  • Connects to a running Resolve via DaVinciResolveScript (Blackmagic’s scripting module at C:\ProgramData\Blackmagic Design\DaVinci Resolve\Support\Developer\Scripting\Modules)
  • Gets the current project -> current timeline (project.GetCurrentProject(), GetCurrentTimeline())
  • Calls project.RenderWithQuickExport("YouTube") - dispatches a Quick Export render
  • On missing method, falls back toe reporting and exits (no full render-settings fallback rigged up.

Significance

  • Confirms the edit lives in DaVinci Resolve (cloud-synced - cf. .syncprojectinfo.json at One-Small-Crawl/.syncprojectinfo.json, ProjectUid 615534cc-75ec-4351-ad13-eeae2138d994).
  • Primary deliverable path: YouTube (the film targets online release -
  • Script sits at the production root with a --help-friendly structure (single use, likely run while Resolve is open with the project loaded.

DaVinci Resolve

Proprietary nonlinear editing, color correction, color grading, VFX, and audio post-production application by Blackmagic Design (Australia). Originally created by da Vinci Systems (USA) and released as da Vinci Resolve in 2004; Blackmagic Design acquired da Vinci Systems in 2009 and continues development.

  • Developer: Blackmagic Design (2009–present); da Vinci Systems (2004–2009).
  • First release: 2004.
  • Stable release: 20.3.2 (12 February 2026).
  • Platforms: macOS, Windows, Linux, iPadOS.
  • Languages: English, Spanish, Japanese, Simplified Chinese, French, Portuguese, Russian, Korean, Thai, Vietnamese.
  • License: Proprietary commercial software, freeware / freemium.
  • Website: blackmagicdesign.com/products/davinciresolve.
  • Task pages: Cut, Edit, Fusion (VFX), Color, Fairlight (audio), Media, Deliver.

Resolve is built around task-specific workspaces (“pages”). The Cut/Edit pages handle video editing, Fusion provides visual effects and motion graphics (previously a standalone app), Color focuses on grading, Fairlight handles audio editing/mixing (previously standalone), Media manages media, and Deliver handles export. Blackmagic markets Resolve as part of a broader peripherals ecosystem (editing keyboards, color grading panels, audio consoles), and Studio is frequently bundled free with Blackmagic camera purchases.

Editions

  • Free – full editing, color, Fusion, Fairlight, delivery. Limited to 4K, 10-bit disabled, no stereoscopic 3D, no HDR grading, no multi-GPU, no neural engine AI features.
  • Studio – paid. Adds resolutions up to 32K, frame rates up to 120 fps, 10-bit video processing, multiple GPU acceleration, stereoscopic 3D, HDR grading, collaborative workflows, additional plug-ins, and AI-driven features.

On One Small Crawl

Resolve is the primary post-production and delivery tool on the production:

  • Cedric Lejeune (cedric-lejeune) runs Resolve-based post and colour work: grading in Blackmagic Resolve, denoising/upscaling of selected shots, ACES (VFX) vs BT.2020 (delivery) debate (2026-07-27-frames-call).
  • 2026-07-20-frames-call, 2026-07-27-frames-call, 2026-08-03-frames-call, 2026-08-10-frames-call, 2026-08-17-frames-call – Resolve referenced across FRAMES calls for grading, animatic assembly, and delivering final shots.
  • render-youtube-final-py (render-youtube-final-py) – a Python script at the production root that drives Resolve’s API (DaVinciResolveScript, Blackmagic’s scripting module at C:\ProgramData\Blackmagic Design\DaVinci Resolve\Support\Developer\Scripting\Modules) via project.GetCurrentProject() / GetCurrentTimeline() and calls RenderWithQuickExport("YouTube") to dispatch a Quick Export render. If the method is missing it falls back to reporting and exits (no full render-settings fallback wired up). This confirms the edit lives in Resolve and that the primary deliverable path is YouTube (online release).
  • The project is cloud-synced: .syncprojectinfo.json at One-Small-Crawl/.syncprojectinfo.json (ProjectUid 615534cc-75ec-4351-ad13-eeae2138d994) ties the Resolve project to One Small Crawl.
  • render-youtube-final-py – Resolve render script (YouTube Quick Export)
  • topaz – Topaz upscaling / enhancement feeding generated and archival frames into Resolve
  • blender – 3D / VFX tool alongside Resolve
  • kitsu – production tracker feeding shot lists into Resolve
  • ai-assisted-post – Topaz / prob4_hyp1 enhancement evidence
  • eddie-ai – Eddie AI, agentic AI editor proposed for rough cuts

Slack

Team collaboration / chat platform used on One Small Crawl for day-to-day communication across the distributed team and work streams. Referenced repeatedly across the FRAMES calls as the operational comms layer.

On One Small Crawl

  • kitsu – production tracker; Kitsu Slack connector was investigated
  • team – the humans + AI agents on the production

Suno

AI music / song generation tool, used on One Small Crawl for the original score direction. Suno generates music (and lyrics) from prompts or references, and is a candidate for producing the film’s score bed.

On One Small Crawl

Topaz (Gigapixel AI / Video AI)

AI-powered image and video enhancement software suite from Topaz Labs (Dallas, Texas). In the One Small Crawl production it is used to upscale and enhance AI-generated and archive frames before they enter the Resolve/composite pipeline.

Topaz Labs develops deep-learning tools for photo and video enhancement: denoising, super-resolution, sharpening, and video upscaling. The company won a Technology & Engineering Emmy Award in 2025 (“AI Image/Video Enhancement for High Quality TV Catalog Restoration”). In June 2026 Adobe announced a definitive agreement to acquire Topaz Labs, expected to close in the second half of 2026. By 2025 Topaz reported ~1.5 million customers and adoption by 20 of the world’s 50 largest companies.

  • Founded: 2006.
  • Headquarters: Dallas, Texas.
  • Flagship products: Photo AI, Video AI, Gigapixel AI, plus cloud apps Bloom, Astra, Express, Mosaic; Topaz Studio bundles desktop + cloud.
  • TIPA 2025: “Enthusiast AI Software”.

Products used on this production

  • Topaz Gigapixel AI – dedicated image upscaling. Enlarges images up to 6x using AI models that synthesize realistic detail rather than interpolating pixels (the first commercially available AI image upscaler, since 2019). Standalone or as a plugin for Photoshop / Lightroom Classic. Nine specialized models: Standard, High Fidelity, Low Res, Text & Shapes, Art & CG, Recover, etc. Gigapixel Pro adds pro-tier models, expanded local processing, 4-image cloud concurrency, full commercial use, seat management.
  • Topaz Video AI – video upscaling, motion deblur, stabilization, grain addition, SDR->HDR conversion. Local + cloud rendering. Rated 4.8 from 3,500+ reviews.
  • Topaz Photo AI – all-in-one denoise + sharpen + upscale + face recovery; plugin for Photoshop, Lightroom Classic, Capture One, Apple Photos, Affinity Photo.

In the FRAMES calls

  • 2026-06-08-frames-call – Topaz surfaced alongside the Gemini shot list (gemini) and the ADI/Eddie AI editor: the shot-blueprint discussion referenced Topaz as the enhancement path for generated frames. Frame upscaling and enhancement evidence lives on ai-assisted-post (Topaz/prob4_hyp1 + AGT-019) and in the raw AI-generated frames note (ai-generated-frames: Topaz TIFFs + prob4_hyp1 clips).
  • Upscaling/enhancement is referenced across several calls (2026-06-01-frames-call, 2026-06-08-frames-call, 2026-07-13-frames-call) as part of the AI-assisted post workflow for cleaning up generated or archival material before compositing in Resolve.

Topaz Labs - image/video AI (no Wikipedia article; official site)

Unreal Engine

Real-time 3D creation engine by Epic Games, used for game development, film/TV production (Virtual Production, nDisplay, Sequencer), architecture, automotive, and simulation. On One Small Crawl, an Unreal Engine 5 project “Desert 5.8” rendered an 802-frame PNG sequence (“DesertSequence”) that may have been used as desert/moon-analog imagery for Scene 4/Scene 5 – though this is a discrepancy with the authoritative pipeline (see unreal-render-pipeline).

  • Developer: Epic Games.
  • First release: 1998 (Unreal Engine 1, for Unreal).
  • Latest generation: Unreal Engine 5 (launched April 2022); UE6 announced May 2026 (Rocket League first confirmed title).
  • Source: available on GitHub; commercial use under royalty model (5% of revenues over US$1M, waived for Epic Games Store exclusives).
  • License: royalty-based commercial; free to use until revenue threshold.
  • Website: unrealengine.com.
  • Notable Guinness record: named “most successful videogame engine” (2014).

Features relevant to this production

  • Sequencer – multi-track editor for cinematic sequences, camera cuts, animation, and render output; the Desert 5.8 PNG sequence likely came from Sequencer/Windows Movie Render Queue.
  • nDisplay – distributed rendering for virtual production / multi-node rendering (used in LED-Volume stages).
  • Lumen / Nanite – dynamic global illumination and virtualized micropolygon geometry (UE5); relevant if high-detail desert/moon assets were involved.
  • Quixel Megascans – photogrammetry/library assets integrated into UE (acquired by Epic); possible source for desert texture/landscape material.
  • Blueprints – visual scripting; possible for pipeline automation, shot export, or render control.
  • Movie Render Queue – high-quality offline rendering path (vs real-time viewport); likely how the 802-frame DesertSequence PNGs were produced.

On One Small Crawl

  • “Desert 5.8” project – an Unreal Engine project at UnrealEngineProjects/Desert 5.8/ rendered an 802-frame PNG sequence DesertSequence.0000.png -> .0801.png (~33s at 24fps) found on disk in Saved/MovieRenders/. This is the only concrete Unreal evidence on the production.
  • Discrepancy – the authoritative pipeline (MovieLabs OMC export, Agent Production Bible v1, era look guide, shot list) describes the pipeline as archive + AI generation + human compositing, with NO Unreal Engine step (generation done by Gen-Image/Gen-Video agents AGT-002/003; the sc04 desert is a consumer-VHS era, not a CG render). See unreal-render-pipeline (low confidence / contested).
  • Likely status – the “Desert 5.8” renders are probably an early/parallel experiment, not part of the authoritative pipeline. Treat as low-confidence background material until confirmed.
  • 2026-07-27-frames-call – snail character versions finished (blend + FBX, tested in Unreal + Blender); gas station environment progressing; Santa Monica managed by alberto-ciprian (2026-07-27-frames-call).
  • 2026-08-03-frames-call – Scene 3 gas-station animatic + AI-generated render shown by Ciprian: started from an Unreal scene; test version with placeholder content (moon landing on the television); Santa Monica blockout also presented with AI-generated buildings; “The Unreal workflow appears here as part of the production pipeline” – see unreal-render-pipeline discrepancy note (2026-08-03-frames-call).
  • 2026-08-10-frames-call – scene blocking in Unreal by gabriele-monteleone; first/last frames extracted for AI generation; shots numbered + reviewed in kitsu (2026-08-10-frames-call).
  • Gabriele Monteleone (gabriele-monteleone) is the DP and Unreal person on the production – scene blocking in Unreal on 2026-08-10-frames-call, scene blocking in Unreal on 2026-08-10-frames-call call (per tim-deussen’s page).
  • Alberto Ciprian (alberto-ciprian) – artist/LoRA/compositing; produced the first snail concept art and Unreal scene renders for Scene 3.

Relationship to authoritative pipeline

The ai-agent-pipeline (ARC retrieval + generation + human compositing) is the authoritative pipeline. The unreal-render-pipeline page documents the discrepancy. If the UE renders are confirmed as an early look-dev for sc04/sc05, this page and the concept page should be moved back to medium confidence and linked to the eras.

  • unreal-render-pipeline – Unreal Render Pipeline (discrepancy) concept page (low confidence)
  • desert-5-8-renders – raw source note: Unreal Engine “Desert 5.8” DesertSequence PNGs
  • blender – Blender, the other 3D tool in use (open-source, snail/blocking work)
  • maya – Maya, industry-standard 3D/animation tool considered alongside Blender
  • ai-agent-pipeline – authoritative pipeline (archive + AI generation + human compositing)
  • era-locked-look – per-scene era emulation (sc04 = consumer-VHS era, not CG)
  • omc-ontology – MovieLabs OMC data model (authoritative pipeline source)
  • gabriele-monteleone – DP, Unreal scene blocking
  • alberto-ciprian – artist/LoRA/compositing, Unreal scene renders for sc03

Unsloth

Unsloth is an open-source inference / fine-tuning tool used to run LLMs efficiently on local hardware. On One Small Crawl it is the inference tool that runs the ornith, lfm and qwen models for hollm.

  • Type: open-source LLM inference / fine-tuning tool.
  • On this production: runs the ornith, lfm and qwen models for hollm on the HP Z2 mini (hp) with the AMD Strix Halo processor (amd).
  • hollm - the assistant that uses Unsloth for inference
  • ornith, lfm, qwen - models run via Unsloth
  • hp, amd - demo hardware

Unsloth - open-source LLM training tool (no Wikipedia article; official site)

Meetings - FRAMES Weekly Calls

Chronological record of the FRAMES IBC2026 Accelerator coordination calls (mostly Monday 4 p.m. weekly calls) that steer one-small-crawl. Source: “Weekly Calls FRAMES.zip” (17 AI-generated meeting-note docx files, ingested into raw/meetings/). Meetings are listed newest-first with headline decisions and links; click a call for full participants, decisions and follow-ups.

Chronology of key decisions (what changed when)

  • 30 Mar: six work streams defined; 3-4 min target output; script is the dependency for ontology/data work (ai-agent-pipeline, omc-ontology).
  • 7 Apr: draft script “The Slow Road” (snail on route-66); MovieLabs proposal merged into FRAMES; archive analysed on-premises for rights.
  • 13 Apr: script widened to space race + vintage cars (archive-led); first snail concept art; lora planned for consistent character (snail, era-locked-look).
  • 20 Apr: Horizon Europe submission; official IBC note-taker only policy; OMC mapping + Moment Labs tagging; script largely approved.
  • 11 May: metadata-analysis workflow (edit-mind + chatgpt) scaled to full dataset; identifier strategy required; agentic-workflow call.
  • 18 May: lora for snail ages done; rai-crits chosen over edit-mind; shot list with rights flags; Streamline demoed; Amici onboard.
  • 1 Jun: kitsunstrated as shot tracker; hardware to Rick; premiere: Rome AI Festival (5 Sep) -> Venice -> IBC; 35mm ambition.
  • 8 Jun: gemini-based shot list with beats/emotions; compliance files auto-updated; ADI (eddie-ai) proposed for rough cuts.
  • 22 Jun: ethics track revived; knowledge graph = single source of truth, kitsu = live view; roles implemented in kitsu.
  • 25 Jun: kitsu + graphdb data flow; AI analysis moved to dailies stage (embeddings, keyframes); agentic automation tools scouted.
  • 29 Jun: transparency framework agreed (C2PA labelling, model cards); shot origin/complexity risk ratings; roles named (director, DP, AI architect, lead AI artist).
  • 13 Jul: IBC marketing/PR plan; ACES vs BT.2020 colour decision; title “One Small Crawl” and snail name “Stella” agreed (supersedes PeriodO “Olivia” - see snail).
  • 20 Jul: five scenes / forty-five shots confirmed; snail rig blend shapes; music by John Hobbs; audio meeting planned; archive matching verified.
  • 27 Jul: archive rights resolved (no third-party sharing); blues-based era-adapted score; “AI generated”/“AI modified” labels agreed.
  • 3 Aug: Scene 3 (gas station) animatic + AI render; first cut with music by 15 Aug; eddie-ai rough-cut comparison; Production AI Card framework.
  • 10 Aug: showcase deadlines locked (deck 4 Sep; showcase 11 Sep, 11:30, Future Tech Stage Hall 14); ethics paper planned; score with suno.
  • 17 Aug: pod logistics/tutorials; google-cloud access granted (Agent Studio, gemini Omni Flash, VO); eddie-ai deep-dive; C2PA export plans.

Call pages (newest first)

2026-08-17-frames-call - 2026-08-17-frames-call

Pod rota + tutorials; google-cloud onboarding (gemini Omni Flash, VO); eddie-ai orchestration (OpenAI/Anthropic/gemini); ethics paper status; all archive footage in resolve; sizzle by 4 Sep.

2026-08-10-frames-call - 2026-08-10-frames-call

Deadlines: POC pass 1 by 21 Aug, deck + cut-down video 4 Sep, pod materials 7 Sep; showcase 11 Sep Hall 14; knowledge-graph framework; archive legal limit (6-7 sequences); ethics slide + paper; suno score.

2026-08-03-frames-call - 2026-08-03-frames-call

Scene 3 gas-station animatic + AI render; Santa Monica blockout; first cut with music by 15 Aug; eddie-ai vs human edit comparison; Kathryn Webb’s Production AI Card framework.

2026-07-27-frames-call - 2026-07-27-frames-call

Locations assigned; master archive rights cleared; blues-based score adapted across eras (one-small-crawl-score); “AI generated” / “AI modified” labels; snail blend+FBX done; Rec.709-first colour with P3D65 final shot.

2026-07-20-frames-call - 2026-07-20-frames-call

5 scenes/45 shots (9 complex) confirmed; post-IBC publication plan; pod tutorial format; archive timecode matching verified; music cues meeting; contributor crediting.

2026-07-13-frames-call - 2026-07-13-frames-call

IBC marketing campaign overview (Jo); kit delivery with HP (hp); ACES chosen for VFX / BT.2020 delivery; snail rig blend shapes; naming: “One Small Crawl” + “Stella” (supersedes PeriodO “Olivia” - see snail); deadlines Aug-Sep.

2026-06-29-frames-call - 2026-06-29-frames-call

Transparency focus: C2PA + risk framework + model cards; EU AI Act labelling (pure/modified/synthetic); MP4/HEVC delivery; roles named; new members Shamir (shamir) and Kathryn (Amici).

2026-06-25-frames-call - 2026-06-25-frames-call

kitsu + graphdb integration; kitsu = source of truth for VFX/comments; dailies-stage AI analysis with embeddings; agent automation tools (OpenClaw, Hermes, NemoClaw). Source filename reads “25.06.25” (typo).

2026-06-22-frames-call - 2026-06-22-frames-call

New partners Amici + eddie-ai; ethics breakout call; HP (hp)/AMD (amd) orders to Rick; knowledge graph = single source of truth; roles into kitsu; gemini shortlist generation with human oversight.

2026-06-08-frames-call - 2026-06-08-frames-call

Hardware deadline; full metadata extraction done; gemini list with emotional beats; topaz licence; Yamdu knowledge-graph import; ADI video editor evaluation; compliance files.

2026-06-01-frames-call - 2026-06-01-frames-call

lora checkpoints ready; kitsu workflow demo (Docker); google-cloud credits; shot categories (archive/semi-automatic/complex); premiere Rome -> Venice -> IBC; Amici governance label.

2026-05-18-frames-call - 2026-05-18-frames-call

lora snail models complete; rai-crits tool selected; shot list with rights/clearance flags; Drive structure aligned to MovieLabs ontology; Streamline platform demo (S3 + Neo4j); Amici intro.

2026-05-11-frames-call - 2026-05-11-frames-call

MPTS spotlight (Clive); metadata analysis to full dataset; identifier strategy (hash vs registry); Scripto/Yamdu ontology pipeline; Friday agentic-workflow call.

2026-04-20-frames-call - 2026-04-20-frames-call

Horizon Europe Cluster 4 submission; official-note-taker-only policy; RAI archive selection; OMC + Moment Labs tagging mapping; C4/handle IDs debate; script near approval.

2026-04-13-frames-call - 2026-04-13-frames-call

Script widened (space race, vintage cars) to fit archive; first snail concept art (3D cartoon, lora plan); RAI research-use rights clarified; external archive outreach (BBC, ITV, ITN, Route 66 museums).

2026-04-07-frames-call - 2026-04-07-frames-call

Contracts + project page logistics; “The Slow Road” draft; archive metadata requirements; on-premises rights-safe analysis; HP (hp) Z-series hardware; MovieLabs merge into FRAMES.

2026-03-30-frames-call - 2026-03-30-frames-call

Project kickoff logistics: six work streams; expertise matrix; script dependency for ontology; sustainability inside responsible-AI; Monday 4 p.m. cadence set.

Cross-cutting decisions log

  • Work streams: Ontology/Metadata, Archive/Capture, Local Agent Orchestration, Post-Production, Responsible AI/Ethics, Creative Story.
  • Output: 3-4 minute piece with 90-second showcase; 35mm version goal.
  • Tools adopted: kitsu (shot tracking, source of truth), Streamline (asset review), Yamdu/Scripto (OMC pipeline), rai-crits (metadata), resolve (edit/grade), topaz (denoise/upscale), eddie-ai (rough cuts; eddie-ai), gemini (shot list + generation), unreal (blocking), suno (score exploration), C2PA (provenance labelling).
  • Provenance/ethics: asset_origin-style categories (archive/hybrid/ synthetic), AI-generated vs AI-modified labels, Production AI Card, C2PA + EU AI Act alignment.
  • Archive reality: RAI masters restricted; ~6-7 sequences cleared; proxies watermarked; use within IBC research only; delete after accelerator.

FRAMES Weekly Call - 2026-08-17

Pod logistics + tutorials, google-cloud access (Agent Studio, gemini Omni Flash, VO) and eddie-ai deep-dive for one-small-crawl.

At a glance

FieldValue
Date2026-08-17-frames-call
Series#17 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 17.08.26.docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions8
Action items5

Participants

NameOrg / Role
roberto-iacoviello-
Muki (muki)-
paola-sunna-
cedric-lejeune-
KathrynAmici
John Canning-
Chris Hampartsoumiangoogle-cloud
Shamir (shamir)-
chris-vienneau-
tim-deussen-

Key decisions

#AreaDecisionFollow-up
1Pod coveragePod rota spreadsheet for IBC coverage; judges may visit incognito; register visitor passes; inform Vanessa for early (Thursday) accessAction 1
2Tutorials~20 minutes each proposed: on-premises workflow, ontology for media creation, AI editor, creativity + agents, post-production workflow, ethics/responsible AI; live demos or screencasts accepted-
3Slide ownersAssigned for the final showcase (AI workflow, post-production, audio, ontology/knowledge graph)Action 2
4google-cloudAccess granted: project members made project owners; Chris Hampartsoumian introduced Agent Studio, gemini Omni Flash (generative video) + VO models (text/image/video-to-video)Action 3
5eddie-aiWorkflow reviewed: ingests footage/scripts/background docs to generate stringouts + rough cuts; orchestrates multiple models (OpenAI, Anthropic, gemini); supports fine-tuning + on-premise deployment; exports to NLEs; C2PA compliance planned; prompt conversations exportable for transparency-
6Edit compareCompare Eddie (eddie-ai)’s automatic edit with Paola (paola-sunna)/Roberto (roberto-iacoviello)’s manual editAction 4
7EthicsPaper progressing (Kathryn): ontology mapped to transparency reporting; model-provider vs deployer legal distinction under the AI Act discussed (err on the side of caution)Action 5
8Post-productionAll archive footage now in resolve; Chris (chris-vienneau) + Cedric (cedric-lejeune) assembling animatics (animatic-workflow) + grades; kitsu migration planned; sizzle video planned for the final PowerPoint by 4 September-

Action items

#ActionOwner
1Add names to the pod rota; email Vanessa for ThursdayAll
2Prepare assigned showcase slides earlyCedric (cedric-lejeune), John Canning, Chris Hampartsoumian, Kathryn
3Contact Chris Hampartsoumian for google-cloud supportAll
4Compare eddie-ai vs manual edits + give feedbackIacoviello (roberto-iacoviello), Paola (paola-sunna)
5Discuss exporting Eddie (eddie-ai) prompt conversationsKathryn, Shamir (shamir)

Notes

FRAMES Weekly Call - 2026-08-10

Final showcase deadlines, technical knowledge-graph framework, creative workflow status + ethics paper plan for one-small-crawl.

At a glance

FieldValue
Date2026-08-10-frames-call
Series#16 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 10.08.26.docx” (Weekly Calls FRAMES.zip)
Participants9
Decisions9
Action items7

Participants

NameOrg / Role
Muki (muki)-
roberto-iacoviello-
chris-vienneau-
paola-sunna-
John-
tim-deussen-
gabriele-monteleone-
Alberto (Ciprian (alberto-ciprian))-
KathrynAmici

Key decisions

#AreaDecisionFollow-up
1TimelinePOC first pass 21 August, second pass 28 August, final showcase deck 4 September, pod materials 7 September; final showcase Friday 11 September, 11:30, Future Tech Stage, Hall 14-
2Pod videoLonger pod video can be finalised up to 9 September; IBC provides the final presentation templateAction 2
3PresentersRoberto (roberto-iacoviello) opens the show (second year), Paola (paola-sunna) + Chris (chris-vienneau) support the presentation-
4FrameworkKnowledge graph as the core; populated with shot lists + archive metadata; tools: kitsu (shot review), resolve (final edit), Streamline (streamline-media) for asset management; provenance chain from render checkpoints to dataset references; every frame traceable to the script-
5Archive limitOnly 6-7 archive sequences available due to legal restrictions; one MXF file, rest are proxies-
6BlockingScene blocking in unreal (unreal-render-pipeline) by Monteleone (gabriele-monteleone); first/last frames extracted for AI generation; shots numbered + reviewed in kitsu-
7MusicJohn developing the score with suno; ethical use of AI music; shared on Streamline (streamline-media) for feedbackAction 7
8EthicsSlide topics: copyright, compensation, fake/malicious content, jobs, sustainability; Kathryn coordinating; collaborative paper planned for publication with the showcaseAction 4
9AI markingOn-screen marking not legally required, but disclosure before viewing may be; practices vary by broadcaster; clarify RAI legal positionAction 6

Action items

#ActionOwner
1Update pod rota spreadsheetAll
2Update showcase presentation with the new IBC templateIacoviello (roberto-iacoviello)
3Check logos + answer slide commentsAll
4Arrange ethics paper drafting meetingKathryn, Muki (muki)
5Add “headache” problem statement + term definitions to the deckIacoviello (roberto-iacoviello)
6Clarify RAI legal policy on AI-generated footage markingIacoviello (roberto-iacoviello)
7Upload first soundtrack stage to Streamline (streamline-media)John

Notes

FRAMES Weekly Call - 2026-08-03

Scene 3 (sc03) animatic (animatic-workflow) + AI render, eddie-ai editing integration and the Production AI Card framework for one-small-crawl.

At a glance

FieldValue
Date2026-08-03-frames-call
Series#15 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 03.08.26.docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions7
Action items5

Participants

NameOrg / Role
Ciprian (alberto-ciprian)-
gabriele-monteleone-
chris-vienneau-
tim-deussen-
John Hobbs (john-hobbs)Composer (soundtrack)
paola-sunna-
roberto-iacoviello-
Shamir (shamir)-
Kathryn WebbAmici
cedric-lejeune-

Key decisions

#AreaDecisionFollow-up
1Scene 3 (sc03)Gas-station animatic (animatic-workflow) + AI-generated render shown by Ciprian (alberto-ciprian): started from an unreal (unreal-render-pipeline) scene; test version with placeholder content (moon landing on the television); Santa Monica blockout also presented with AI-generated buildings-
2FeedbackPush realism, add environmental effects, refine AI artefacts; after-effects used for transitions + old-movie effects-
3ConsistencyAll environments to be brought to a consistent level for camera composition + silhouette; all shots in kitsu + emotional beats agreed before Gabriela’s vacation-
4Cut targetFirst cut with music by 15 August (vacation milestone); iterate animation + visual quality in the second half of August-
5Music needsJohn needs a mood timeline + scene timings to start music editing; music will guide the editAction 1
6eddie-aiWorkflow: ingest footage from Google Drive, generate a rough cut, export to resolve; separate resolve Cloud project for AI edits to avoid risk to the main project; compare Eddie output with human edits (Paola (paola-sunna)/Roberto (roberto-iacoviello))Action 3
7AI CardProduction AI Card framework presented by Kathryn Webb: documents AI use across production stages, distinguishes visible vs invisible AI; encourages good practice but is not a substitute for legal sign-off; team to answer framework questionsAction 5

Action items

#ActionOwner
1Send mood timeline + scene timings to JohnChris (chris-vienneau)
2Document learnings/time for the model buildCiprian (alberto-ciprian)
3Send archive footage, script + notes to Shamir (shamir) for the Eddie (eddie-ai) testPaola (paola-sunna), Roberto (roberto-iacoviello)
4Share vocabulary/ontology definitions with KathrynChris (chris-vienneau)
5Contribute answers to AI Card questions in slackPaola (paola-sunna), team

Notes

FRAMES Weekly Call - 2026-07-27

Location assignments, archive rights resolution, colour pipeline and blues-based era-adapted score for one-small-crawl.

At a glance

FieldValue
Date2026-07-27-frames-call
Series#14 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 27.07.26.docx” (Weekly Calls FRAMES.zip)
Participants11
Decisions12
Action items6

Participants

NameOrg / Role
Muki (muki)-
roberto-iacoviello-
paola-sunna-
chris-vienneau-
tim-deussen-
John Hobbs (john-hobbs)Composer (soundtrack)
Ben-
KathrynAmici
Ciprian (alberto-ciprian)-
Gabriel-
cedric-lejeune-

Key decisions

#AreaDecisionFollow-up
1LocationsLocation assignments for project scenes completed-
2google-cloudAccess list sent to Chris (Google) but he was on holiday; Muki (muki) to nudge; access needed for VO 3.1 + kitsuAction 1
3HardwareAMD (amd) machines not yet received; Rick (hp) shipped the required laptop-
4Shot listYamdu shot list is a direct copy of Tim (tim-deussen)’s Excel list, the master reference for analysis + shot numbering-
5Archive matchMatch broadly to story requirements rather than regenerate the shot list for perfect matches-
6Archive rightsMaster archive access secured; rights issues resolved; usage protocol: no third-party sharing, deletion after the accelerator periodAction 5
7StoryboardsTwo storyboards (kitsu + Chris (chris-vienneau)’s software) are placeholders; focus on integrating music + shots for editing-
8SoundtrackSingle piece based on a traditional blues chord sequence, adapted in style + sound quality across the eras; dialogue mainly from archive, Foley from library + possibly generative AI; John’s music as the overarching score. Scene timings available as EDL or AAFAction 6
9Ethics labellingTwo labels - “AI generated” + “AI modified”; track human involvement (tool selection, approvals) for copyrightAction 4
10Character/envSnail character versions finished (blend + FBX, tested in unreal (unreal-render-pipeline) + blender); gas station environment progressing; Santa Monica managed by Ciprian (alberto-ciprian)Action 3?
11ColourFootage stays in original colour space (mostly Rec.709); final shot mastered in P3D65 within a BT.2020 container; Gen AI footage conformed to archive colour space + gamma-
12PlayboxProposed transcoding raw files + building a promotional playlist (pending group approval)-

Action items

#ActionOwner
1Nudge Chris (chris-vienneau) re: google-cloud accessMuki (muki)
2Securely share one MXF clip with Cedric (cedric-lejeune) + Chris (chris-vienneau) for colour testsIacoviello (roberto-iacoviello)
3Review + tag archive material on Streamline (streamline-media)Paola (paola-sunna), Iacoviello (roberto-iacoviello)
4Poll team for the AI ethics breakout callIacoviello (roberto-iacoviello)
5Circulate archive rights recommendationsIacoviello (roberto-iacoviello)
6Share Google Drive links with BenPaola (paola-sunna), Iacoviello (roberto-iacoviello)

Notes

FRAMES Weekly Call - 2026-07-20

IBC presentation planning, shot/archive coordination and music production for one-small-crawl.

At a glance

FieldValue
Date2026-07-20-frames-call
Series#13 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 20.07.26.docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions8
Action items7

Participants

NameOrg / Role
Muki (muki)-
chris-vienneau-
tim-deussen-
roberto-iacoviello-
paola-sunna-
Florian-
John Hobbs (john-hobbs)Composer (soundtrack)
Gabrielle-
alberto-ciprian-
gabriele-monteleone-

Key decisions

#AreaDecisionFollow-up
1PublicationPost-IBC publication confirmed: agreed presentation materials published on the IBC website + reusable at other conferences (no embargo); track opportunities in a shared scheduler spreadsheet; credit contributors (e.g. Alberto (alberto-ciprian)) when work is presentedAction 1
2POC materialsFor IBC: sizzle videos, under-the-hood architecture, colourful visuals, wall of learnings, ethics paperAction 2
3Pod formatRepeat last year’s approach: 2 tutorials/day featuring software demos; pod hardware lockable, HDMI, supported by IBC staff; demo setup testable the day before-
4StatusFive scenes + forty-five shots, nine complex; Scene 4 (sc04) most shot-dense, Scene 3 (sc03) most complex VFX-wise-
5CharacterSnail confirmed as“Stella“; Chris (chris-vienneau)’s daughter provides voice + sound effects; John Hobbs (john-hobbs) composing the soundtrack; Alberto (alberto-ciprian) rigging three character versions with emotional blend shapes (conflict with “Olivia” - see snail)Action 4
6Archive verificationTimecodes from Tim (tim-deussen)’s Excel matched AI-generated descriptions + videos for most shots; missing videos + naming errors remain; human review required for footage selectionAction 7
7Metadata/provenanceVia Streamline (streamline-media) work panel tasks per shot; moving away from spreadsheets; alternative archive matches per shot requestedAction 6
8AudioSingle composed score adapted per era; dedicated audio meeting for cues; human-driven music creation (no generic AI tracks)Action 3

Action items

#ActionOwner
1Create conference-opportunity spreadsheet tabIacoviello (roberto-iacoviello), Paola (paola-sunna)
2Compile proof-of-concept checklist with ownersIacoviello (roberto-iacoviello), Paola (paola-sunna)
3Schedule audio meeting; provide cues, script, timeline to JohnChris (chris-vienneau)
4Upload emotional blend shapes to kitsu for reviewMonteleone (gabriele-monteleone)
5Share naming convention recommendationFlorian
6Burn timecodes, upload footage, create tasks in Streamline (streamline-media) per shotMonteleone (gabriele-monteleone), John, Paola (paola-sunna), Iacoviello (roberto-iacoviello)
7Relaunch selection script to output multiple archive variants per shotTim (tim-deussen)

Notes

FRAMES Weekly Call - 2026-07-13

IBC marketing/PR plan, technical pipeline review (colour, rig) and naming agreement for one-small-crawl.

At a glance

FieldValue
Date2026-07-13-frames-call
Series#12 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 13.07.26.docx” (Weekly Calls FRAMES.zip)
Participants11
Decisions9
Action items6

Participants

NameOrg / Role
JoIBC marketing
Muki (muki)-
Shamir (shamir)-
roberto-iacoviello-
cedric-lejeune-
John-
chris-vienneau-
tim-deussen-
gabriele-monteleone-
Gabrielle-
alberto-ciprian-

Key decisions

#AreaDecisionFollow-up
1MarketingJo presented the IBC2026 Accelerators campaign: Future Tech innovation destination, Accelerators Zone at the heart, EIT innovation partner, sponsors Google/HP (hp)/Shaw; Kickstart Day in February; 200k+ contact database; IBC365 editorial; per-project marketing toolkits; IBC Exchange Pass (5 per project) for the Brain Date Lounge-
2ExhibitAll projects have a presence in Hall 14; companies can buy separate stands at extra cost; pod rota system for staffing; teams should represent the collaborative project, not company promo-
3Kit deliveryCheck with Rick Holman (hp) on AMD (amd)/HP (hp) kit deliveriesAction 1
4Archive workflowWatermarked proxies for review on Streamline (streamline-media); masters restricted; Cedric (cedric-lejeune) assists with denoising/upscaling selected shotsActions 3-4
5Colourresolve with Cedric (cedric-lejeune) grading; ACES vs BT.2020 debated: ACES more flexible for VFX, BT.2020 for delivery-
6RigRoberto (roberto-iacoviello) showed the snail rig in blender: eye controls + blend shapes for emotions; young + adult versions finalised; templates for preset emotions-
7Scale/arcScale + emotional arc of the snail to be finalised in a separate review (Tim (tim-deussen) + Gabrielle); concept art uploaded to the shared drive-
8DeliverablesFirst pass review last week of August; final assets due 4 September for the stage presentation; Friday showcase slot requested (as in the previous year)Action 6
9NamingFilm titled “One Small Crawl” and the snail named“Stella“; update kitsu, Streamline (streamline-media) + all documentation (conflict with the PeriodO crosswalk “Olivia” - see snail)Action 5

Action items

#ActionOwner
1Check AMD (amd)/HP (hp) delivery status with RickRoberto (roberto-iacoviello)
2Delete duplicate project on Streamline (streamline-media) + confirm upload process for archive footageIacoviello (roberto-iacoviello)
3Upload watermarked proxies to Streamline (streamline-media); give Cedric (cedric-lejeune) selected clean shotsIacoviello (roberto-iacoviello)
4Provide master archive example to Cedric (cedric-lejeune) + Chris (chris-vienneau)Roberto (roberto-iacoviello)
5Update platforms to “One Small Crawl” / “Stella”Iacoviello (roberto-iacoviello)
6Book Friday showcase slot with Muki (muki)Iacoviello (roberto-iacoviello), Muki (muki)

Notes

FRAMES Weekly Call - 2026-06-29

Transparency framework (C2PA), production workflow, role assignment + new members for one-small-crawl.

At a glance

FieldValue
Date2026-06-29-frames-call
Series#11 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 29.06.26 (1).docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions8
Action items6

Participants

NameOrg / Role
Muki (muki)-
roberto-iacoviello-
John-
AhsanAmici
KathrynAmici
Shamir (shamir)-
chris-vienneau-
tim-deussen-
Erik-
Florian-

Key decisions

#AreaDecisionFollow-up
1MarketingJo’s marketing campaign overview rescheduled to next week (wrong call)Action 1
2TransparencyResponsible AI narrowed to transparency: C2PA content labelling, risk-adjusted framework + decision models for AI tool selection, aligned with EU AI Act; audience labels for pure, modified + synthetic contentAction 3
3Model cardsRequired for underlying AI tools; Amici to provide a label for responsible AI-
4C2PA techTechnical options discussed (frame-level metadata, server-side certificates) alongside emerging European parallel standards-
5FormatsDistribution: MP4 + HEVC; production: ProRes or XAVC; asset management tracks versions + provenance-
6Shot breakdownBy origin (archive, hybrid, synthetic) + complexity presented with risk ratings, aligned with EU AI Act labellingAction 6
7RolesStudio Dyson as director, Gabriele Monteleone (gabriele-monteleone) as DP, Chris (chris-vienneau) Vienneau as AI architect, Iacoviello (roberto-iacoviello) as lead AI artist, others for AI artistry, asset management, metadata-
8New membersShamir (shamir) introduced; Kathryn joined from Amici-

Action items

#ActionOwner
1Coordinate with Jo for the marketing overview next weekMuki (muki)
2Insert proof-of-concept deadlines (deck, sizzle) into the timeline for late-August reviewMuki (muki), Iacoviello (roberto-iacoviello)
3Schedule the transparency focus call within Friday’s meetingAhsan, Iacoviello (roberto-iacoviello), Kathryn
4Grant Ahsan access to the FRAMES IBC 2026 Google DriveMuki (muki)
5Shamir (shamir) to assist with C2PA implementation questionsShamir (shamir)
6Share shot origin/complexity/risk chartsChris (chris-vienneau)

Notes

FRAMES Weekly Call - 2026-06-25

kitsu setup, graph-database data flow and dailies-stage AI analysis for one-small-crawl.

Source filename reads “25.06.25”; dated 2026-06-25-frames-call from content + position between the 22.06 and 29.06 calls (filename year typo).

At a glance

FieldValue
Date2026-06-25-frames-call
Series#10 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 25.06.25 (1).docx” (Weekly Calls FRAMES.zip)
Participants4
Decisions6
Action items3

Participants

NameOrg / Role
cedric-lejeune-
chris-vienneau-
roberto-iacoviello-
tim-deussen-

Key decisions

#AreaDecisionFollow-up
1kitsuEnvironment set up by Cedric (cedric-lejeune): manages sequences, shots, reviews; H264 player playback; comments on shots (no drawing); plugin architecture can add custom fields (e.g. prompts in description)Actions 1-2
2DeploymentLocal via Docker or cloud; some services (graph database) may be better suited to cloud hosting-
3graphdbSet up with Docker; hosted locally or on Azure; imports/merges data from Scripto, FilmUStage, Scriptation, Yamdu with custom fields + relationships (creative works, scenes, characters, locations)-
4Truth sourceskitsu is the source of truth for VFX, comments + status; the graph database stores metadata with backup + versioning-
5Video pipelineAI analysis shifted to the dailies stage (low-res video for keyframes, face detection); vector embeddings for search (open source, e.g. Nomic); full version history maintained-
6AutomationAgents to update kitsu with new renders from Google Drive; tools discussed include OpenClaw, Hermes, NemoClaw for secure agentic orchestrationAction 3

Action items

#ActionOwner
1Send kitsu link to Chris (chris-vienneau); test import/exportCedric (cedric-lejeune)
2Check for a kitsu connectorCedric (cedric-lejeune)
3Investigate NemoClaw integration into the server environmentCedric (cedric-lejeune)

Notes

FRAMES Weekly Call - 2026-06-22

Partner logos, ethics track revival, hardware orders and role attribution for one-small-crawl.

At a glance

FieldValue
Date2026-06-22-frames-call
Series#9 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 22.06.26.docx” (Weekly Calls FRAMES.zip)
Participants8
Decisions10
Action items5

Participants

NameOrg / Role
Muki (muki)-
paola-sunna-
cedric-lejeune-
John Hobbs (john-hobbs)Streamline (streamline-media)
John Canning-
roberto-iacoviello-
tim-deussen-
Erik-

Key decisions

#AreaDecisionFollow-up
1LogosCedric (cedric-lejeune) + John Hobbs (john-hobbs) sent university logos; Paola (paola-sunna) to send the EBU logo after checking with a colleagueAction 1
2PartnersNew partners confirmed: Amici + eddie-ai; ensure Shamir Allibhai (shamir) is on all communications; follow up with maya (Playbox)Action 5
3EthicsTrack revived; breakout call planned (Wednesday) to clarify partner contributions + request governance/transparency guidelines from Amici + eddie-aiAction 3
4HardwareAll orders (HP (hp)/AMD (amd)) consolidated + sent to Rick Pullman; Iacoviello (roberto-iacoviello) responsible; order not yet submitted (deadline passed)Action 2
5Agent demoCedric (cedric-lejeune) to run an agent for image generation on a local laptop for a DMC days demo-
6ShortlistingIteration process: archive footage + creative intent, using MovieLabs ontology (omc-ontology) + google-cloud gemini API with multiple agents; human oversight essential for final shot selection-
7LegalQuestions on synthetic/composite shots + EU labelling routed to Amici-
8google-cloudVirtual machine + 1 TB storage confirmed; G Drive for full-resolution files-
9RolesAttributed for the production (executive producer, AI architect, VFX artists, etc.) + implemented in kitsuAction 4
10Truth sourcesKnowledge graph = single source of truth; kitsu = live organisational view; kitsu MCP under implementation for agent management-

Action items

#ActionOwner
1Send EBU logo to OctaviaPaola (paola-sunna)
2Submit consolidated HP (hp)/AMD (amd) hardware order, copying John Canning, Muki (muki), Paola (paola-sunna)Iacoviello (roberto-iacoviello)
3Request governance/transparency guidelines from Amici + eddie-aiPaola (paola-sunna)
4Finalise role list for kitsuCedric (cedric-lejeune)
5Ensure Shamir Allibhai (shamir) added to groupMuki (muki)

Notes

FRAMES Weekly Call - 2026-06-08

Hardware order deadline, gemini-driven shot list analysis and compliance documentation for one-small-crawl.

At a glance

FieldValue
Date2026-06-08-frames-call
Series#8 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 08.06.26 (1).docx” (Weekly Calls FRAMES.zip)
Participants11
Decisions8
Action items6

Participants

NameOrg / Role
Muki (muki)-
paola-sunna-
roberto-iacoviello-
tim-deussen-
chris-vienneau-
gabriele-monteleone-
cedric-lejeune-
aidan-enigmatic-
davide-zappia-
John-
Florian-

Key decisions

#AreaDecisionFollow-up
1HardwareAMD (amd)/HP (hp) hardware order must be submitted before the Friday deadline; copy John + Muki (muki) for oversightAction 2
2MetadataExtraction completed on all clips (Ray); Roberto (roberto-iacoviello) downloaded the “100 Years of Television” footageAction 4
3Shot list appTim (tim-deussen)’s app (gemini Pro API + MovieLabs ontology (omc-ontology)) produces beat summaries, emotional shifts, shot types, transitions, camera angle, duration; visualised in Excel + a timeline tool-
4Knowledge graphShot list to be imported into Yamdu, generating a knowledge graph with stable identifiers (EU-compliant UUIDs) for kitsu + other systems-
5Monteleone (gabriele-monteleone) feedbackSome AI-chosen shots (e.g. title sequences) are not suitable for compositing; alternative rankings + metadata marking needed-
6topazAI Video denoising experiments by Aidan (aidan-enigmatic) need a full licence; Chris (chris-vienneau)/John to facilitate; resolve considered as fallback for upscaling hallucinationsAction 3
7ComplianceFiles (AI Act, copyright, data protection, sustainability) auto-updated during development; transparency audits + provenance linked to the knowledge graph-
8eddie-aiAssistant video editor proposed for automated rough cuts from the shot list; follow-up call with the ADI co-founder on FridayAction 6

Action items

#ActionOwner
1Send logos to OctaviaAll
2Finalise AMD (amd)/HP (hp) hardware order, copying John Canning + Muki (muki)Paola (paola-sunna), Roberto (roberto-iacoviello)
3Obtain topaz AI Video licence; upload comparison resultsChris (chris-vienneau), John; Aidan (aidan-enigmatic)
4Add 100 Years of Television footage to metadata analysisTim (tim-deussen)
5Share ETC/Disney ontology + story beat definitions with Tim (tim-deussen)Chris (chris-vienneau)
6Evaluate ADI video editor on the Friday callAll

Notes

FRAMES Weekly Call - 2026-06-01

lora completion, kitsu workflow demonstration, google-cloud accessand premiere for one-small-crawl.

At a glance

FieldValue
Date2026-06-01-frames-call
Series#7 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 01.06.26 (1).docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions10
Action items5

Participants

NameOrg / Role
Muki (muki)-
roberto-iacoviello-
chris-vienneau-
tim-deussen-
cedric-lejeune-
Erik-
AhsanAmici
alberto-ciprian-
aidan-enigmatic-
David-

Key decisions

#AreaDecisionFollow-up
1loraAlberto (alberto-ciprian) completed loraing of the snail character across ages; checkpoints available on Google Drive-
2MetadataExtraction ongoing (managed by another R&D group); Tim (tim-deussen) awaits it to synchronise script + archive material for the shot list-
3CompositingSnail + archive compositing workflow in development by Alberto (alberto-ciprian); David working on super-resolution upscaling hallucinations-
4HardwareRequests consolidated + submitted to Rick by 12 JuneAction 1
5Enigmatic/Cedric (cedric-lejeune)Aidan (aidan-enigmatic) working on denoising, inpainting, outpainting; Cedric setting up the kitsu workflow for project management-
6google-cloudAccess to be requested from Chris H (Google); no strict credit limit - shared allocation across all acceleratorsAction 2
7kitsuCedric (cedric-lejeune) demoed (open-source project tracker: sequences, shots, assets, tasks; local deployment via Docker possible)-
8Shot categorisationOnce the shot list is final, shots categorised (archive, semi-automatic, complex), assigned in kitsu, tracked with progress reviews-
9AmiciAmici (Ahsan) to provide thought leadership on AI governance and potentially develop a responsible-AI labelAction 4
10premiereRome AI Festival on 5 September, then Venice + IBC; 35mm version expected for screening-

Action items

#ActionOwner
1Compile hardware request list to RickIacoviello (roberto-iacoviello), Paola (paola-sunna)
2Email Chris H at Google for google-cloud credentials, copying Muki (muki) + AbbyIacoviello (roberto-iacoviello)
3Distribute MovieLabs creative vocabulary to the teamChris (chris-vienneau)
4Prepare ethics/responsibility presentation with CatherineAhsan
5Maintain shared learnings documentPaola (paola-sunna)

Notes

FRAMES Weekly Call - 2026-05-18

lora model completion, rai-crits tool selection, shot list generation and Streamline (streamline-media)/Amici onboarding for one-small-crawl.

At a glance

FieldValue
Date2026-05-18-frames-call
Series#6 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 18.05.26.docx” (Weekly Calls FRAMES.zip)
Participants8
Decisions8
Action items5

Participants

NameOrg / Role
alberto-ciprian-
roberto-iacoviello-
chris-vienneau-
davide-zappia-
tim-deussen-
John Hobbs (john-hobbs)Streamline (streamline-media)
Muki (muki)-
Ahsan MallickAmici

Key decisions

#AreaDecisionFollow-up
1loralora for the snail character at different ages trained successfully by Ciprian (alberto-ciprian); test prompts + parameters on Google Drive; compositing tests started, facial expressions need improvement-
2TraceabilityTrained images to be tagged + linked to the snail character in Yamdu for traceability (naming-schema, provenance)-
3Tool selectionrai-crits selected over edit-mind for metadata extraction: richer semantic output (audio transcription with speaker diarization, named entity recognition, scene analysis), shot-block granularity, human-readable IDs, MAM integration-
4Shot listTim (tim-deussen) produced a shot list mapping archive footage to script locations with timecodes + production notes; flags for archive clearance, brand/logo, likeness, music, + human validation of AI metadataAction 1
5Drive structureGoogle Drive folder structure + workflow documentation aligned to the MovieLabs ontology (omc-ontology); JSON, JSON-LD + TTL exports carry scene + character identifiersAction 2
6Streamline (streamline-media)John Hobbs (john-hobbs) demoed: AWS S3 + Neo4j graph database, versioning by humans + machines, comments as first-class assets, M2M API for agents; positioned as alternative to Frame.io-
7AmiciAhsan Mallick introduced Amici: AI governance platform for film/TV, launched an AI competency course based on the Alan Turing framework-
8Best practicesIdentifier best-practices folder planned for interoperability-

Action items

#ActionOwner
1Formally request Yamdu API support for shot list importTim (tim-deussen)
2Send MovieLabs RDF/ontology data + versioning spec to JohnChris (chris-vienneau)
3Onboard Real Dreams to the acceleratorMuki (muki), Iacoviello (roberto-iacoviello)
4Upload further archive metadata analyses to shared driveZappia (davide-zappia)
5Commit to AI workflow tasks in the shared listAll

Notes

FRAMES Weekly Call - 2026-05-11

MPTS spotlight, metadata-analysis workflow demonstration and identifier strategy for one-small-crawl.

At a glance

FieldValue
Date2026-05-11-frames-call
Series#5 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 11.05.26.docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions7
Action items5

Participants

NameOrg / Role
Muki (muki)-
Clive Santa Maria-
tim-deussen-
paola-sunna-
roberto-iacoviello-
David (Zappia (davide-zappia))-
chris-vienneau-
Raymond-
Erik-
alberto-ciprian-

Key decisions

#AreaDecisionFollow-up
1SpotlightProject featured in the MPTS (London) accelerator spotlight; Clive presents the work streams on stageAction 1
2Metadata workflowDemoed by David (davide-zappia): JSON from archive material with edit-mind, analysed with chatgpt to extract keywords, ontology classes + a concept document; mapped to script sections (tools: YOLO, Deepface, Qwen 2.5)Actions 3-4
3Scale-upExtend metadata analysis to the full archive video dataset, pending approvalAction 3
4IdentifiersPersistent unique identifier strategy required: hash-based (mathematical uniqueness) vs registry/file-system-based debated; group to define itAction 2
5OMC pipelineMovieLabs ontology implementation on track: Scripto for writing, Yamdu for planning; goal of stable scenes/characters/locations + MovieLabs JSON export (omc-ontology)-
6Agentic workflowDedicated agentic workflow call scheduled (Friday) with Enigmatic, ITV et al.; priorities: prompt enhancement, content announcement, media generationAction 5
7loraloraing for concept art restarted after machine issues (Ciprian (alberto-ciprian))-

Action items

#ActionOwner
1Submit logos + signed contracts for MPTS slidesAll
2Develop + agree the identifier strategyChris (chris-vienneau), Erik, Raymond, Iacoviello (roberto-iacoviello), David (davide-zappia)
3Apply metadata workflow to all archive videosTim (tim-deussen), Iacoviello (roberto-iacoviello)
4Clarify frame-level vs sampled metadata extractionDavid (davide-zappia)
5Send invitations for the agentic workflow callIacoviello (roberto-iacoviello)

Notes

FRAMES Weekly Call - 2026-04-20

EU funding status, AI note-taker policy, project overview and OMC ontology (omc-ontology) integration for one-small-crawl.

At a glance

FieldValue
Date2026-04-20-frames-call
Series#4 of 17 FRAMES weekly calls
TypeMeeting Notes
Source“Meeting Notes 20.04.26.docx” (Weekly Calls FRAMES.zip)
Participants8
Decisions7
Action items7

Participants

NameOrg / Role
roberto-iacoviello-
Clive Santa Maria-
tim-deussen-
Muki (muki)-
John-
Raymond-
paola-sunna-
cedric-lejeune-

Key decisions

#AreaDecisionFollow-up
1EU fundingHorizon Europe Cluster 4 proposal submitted; only 6-7 of 107 proposals funded, highly competitive; team joins as invited partner, not leader-
2Note-taker policyOnly the official IBC note-taker allowed on calls; unauthorised tools (Otter AI, Fireflies) removed for security and complianceAction 1
3OverviewAI agent-based workflow, snail journey along route-66, archive + generated content, era-specific visual styles (b/w for analogue, colour for digital), minimal dialogue-
4Archive selectionRAI selection presented: route-66, radio/TV evolution, space race; work continues on tagging, indexing, integration-
5Ontology mappingMap scripts + tagged content to the OMC ontology (omc-ontology); Moment Labs tagging mapped via scripts; C4 IDs vs handle systems debated for asset identificationAction 6
6EthicsDedicated ethics + responsible AI stream maintained; legal questions (compositing AI material with real footage vs using archive as training data) escalated to legalAction 7
7Script statusLargely approved; final sign-off after remaining stakeholder input, deadline end of weekAction 5

Action items

#ActionOwner
1Remove unauthorised AI note-takers; official IBC note-taker onlyMuki (muki)
2Complete IBC contractsAll
3Post agenda in slack thread before each callPaola (paola-sunna), Iacoviello (roberto-iacoviello)
4Send John the master Google Drive + PM spreadsheet linksIacoviello (roberto-iacoviello), Paola (paola-sunna)
5Send script link to John + Eric for reviewIacoviello (roberto-iacoviello), Paola (paola-sunna)
6Arrange Moment Labs call re: tagging + data exportPaola (paola-sunna), Iacoviello (roberto-iacoviello)
7Legal guidance on AI + archive footage distinctionIacoviello (roberto-iacoviello)

Notes

FRAMES Weekly Call - 2026-04-13

Script update (widened scope), first snail concept art, lora plan and archive rights strategy for one-small-crawl.

At a glance

FieldValue
Date2026-04-13-frames-call
Series#3 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 13.04.26 (1).docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions7
Action items6

Participants

NameOrg / Role
roberto-iacoviello-
gabriele-monteleone-
Muki (muki)-
paola-sunna-
tim-deussen-
alberto-ciprian-
chris-vienneau-
Nicolas-
Clive Santa Maria-
John-

Key decisions

#AreaDecisionFollow-up
1Script scopeWidened to a broader definition of technology (space race, vintage cars) because relevant RAI archive footage on TV/cinema evolution was limited-
2Script approvalApproval is critical: storyboard + character design are blocked until final; the week is treated as a creative sprint-
3Concept artFirst snail concept art by Ciprian (alberto-ciprian): 3D model, cartoonish, non-photorealistic; lora (Low-Rank Adaptation) proposed to train the character for consistent animation across eras-
4Archive rightsRAI material usable for research within the IBC group; public/scientific use requires explicit authorization; fine-tuning AI models on the content needs checkingAction 2-3
5External archivesReach out to BBC, ITV, ITN, Route 66 museums; moments-lab considered for content retrieval + semantic analysisAction 4
6Working editClive: create a working edit with available footage and replace shots as rights are cleared, to avoid delays-
7StreamsWork stream leads confirmed (e.g. Tim (tim-deussen) and Eric for creative story); dedicated slack per stream-

Action items

#ActionOwner
1Review script on Google Drive + comment; slack poll for approvalAll, Iacoviello (roberto-iacoviello)
2Download archive content + share with Tim (tim-deussen) and EricIacoviello (roberto-iacoviello)
3Confirm whether fine-tuning on archive material is permittedIacoviello (roberto-iacoviello)
4Contact route-66 museum (Illinois) + Chicago museum for cleared footageMuki (muki)
5Check Jo Mayer re: last year’s IBC exhibit footage rightsMuki (muki)
6Move archive footage discussions to Work Stream 2 (archive retrieval) slack channelIacoviello (roberto-iacoviello), Paola (paola-sunna), Clive, Muki (muki)

Notes

FRAMES Weekly Call - 2026-04-07

Contract onboarding, first script draft (“The Slow Road”), archive metadata requirements and technical/hardware setup for one-small-crawl.

At a glance

FieldValue
Date2026-04-07-frames-call
Series#2 of 17 FRAMES weekly calls
TypeWeekly Call
Source“Weekly Call 07.04.26 (1).docx” (Weekly Calls FRAMES.zip)
Participants10
Decisions8
Action items5

Participants

NameOrg / Role
Muki (muki)-
Florian-
roberto-iacoviello-
paola-sunna-
chris-vienneau-
cedric-lejeune-
tim-deussen-
maya-
John Canning-
Erik-

Key decisions

#AreaDecisionFollow-up
1PublishingProject page to be published Thursday/Friday; contracts must be signed before company names and logos appear on it-
2ScriptDraft titled “The Slow Road”: a cartoon snail travelling route-66 as a metaphor for the history of television; AI-generated, needs human revision for depth + memorable ending-
3Archive metadataArchive colleagues must supply detailed production metadata (camera types, film stock, production documents) to authenticate recreated footage + support AI workflowsAction 2
4Parallel tracksCreative development + archive research run in parallel; narrative adapts to available archival material + rights-
5MovieLabsProposal merged into the main project; work streams and shared OMC ontology integrated (omc-ontology)-
6ArchivesAnalysed on-premises to avoid rights issues; AI extracts style/inspiration without transferring originals offsite-
7HardwareHP (hp) Z-series workstations discussed; Erik shared a reference architecture (NetBird + Create) for unified local/cloud env; pre-configuration for less technical participantsAction 4
8IBC resource“100 Years of Television” compilation requested from IBC as potential archive resource-

Action items

#ActionOwner
1Email tier level to Muki (muki) if contract not receivedAll
2Collect archive metadata + production infoIacoviello (roberto-iacoviello)
3Update expertise/contribution matrix; add hardware requirementsAll participants
4Share reference architecture documentErik
5Send high-res logos to OctaviaAll

Notes

FRAMES Weekly Call - 2026-03-30

Project kickoff: work-stream structure, output targets and cadence for the FRAMES accelerator hosting one-small-crawl.

At a glance

FieldValue
Date2026-03-30-frames-call
Series#1 of 17 FRAMES weekly calls
TypeMeeting Notes
Source“Meeting Notes 30.03.26 (1).docx” (Weekly Calls FRAMES.zip)
Participants14
Decisions7
Action items6

Participants

NameOrg / Role
Muki (muki)IBC coordinator
OctaviaIBC marketing
paola-sunna-
roberto-iacoviello-
maya Ash-
Slavi-
chris-vienneau-
Clive Santa Maria-
Raymond-
Nicolas-
Erik-
tim-deussen-
cedric-lejeune-
John Canning-

Key decisions

#AreaDecisionFollow-up
1Work streamsSix streams defined (Ontology & Metadata, Archive & Capture, Local Agent Orchestration, Post-Production, Responsible AI/Ethics, Creative Story); milestones for July and September-
2OutputFinal output 3-4 min with a 90 s showcase; a 30 s advert debated but flagged as splitting focus-
3SustainabilityStays inside the Responsible AI stream; explainability + risk assessment prioritized first-
4ScriptDocumentary/docu-fiction genre; script is a prerequisite for ontology and data work; 2-week target (mid-April)Action 5
5PrototypingRapid prototyping (techvis) + animatics (animatic-workflow) start in parallel, before script is final, to iterate with archive content-
6ToolsExpert AI (Italian partner) offers its knowledge-graph product free; Moment Labs agentic layer on Mega DBU considered for archive integrationAction 6
7CadenceRegular Monday 4 p.m. team calls; IBC Accelerators happy hour at NAB (Las Vegas) and MPTS (London) as milestones-

Action items

#ActionOwner
1Send high-res JPEG company logos to OctaviaAll
2Follow up contracts for Paola (paola-sunna), maya Ash et al.Muki (muki)
3Onboard maya Ash + Slavi into slackMuki (muki)
4Build expertise/contribution matrix to identify skill gapsIacoviello (roberto-iacoviello), Clive, Muki (muki)
5Draft new script within two weeksPaola (paola-sunna), Iacoviello (roberto-iacoviello)
6Contact expert AI partner, overcome language barriersmaya Ash, Paola (paola-sunna), Iacoviello (roberto-iacoviello)

Notes

AI Accountability

The question of who is liable for due diligence and risk assessment when something goes wrong with AI use in a production, and at what stage of the production this should be considered.

Key questions raised (from the paper)

  • Who is liable for due diligence and risk assessment, especially if something goes wrong? The prompt engineer? The production company? The technical team? The AI model provider or the creator platform?
  • At what stage of the production should this be considered?
  • What exactly are the guarantees of legal indemnities (contractual safeguards against copyright infringement lawsuits from generative AI output or training data)? Under what conditions do such indemnities become void?

FRAMES context

  • Accountability is one of the responsible-AI themes (see responsible-ai) alongside copyright, transparency and disclosure.
  • The paper’s closing argument (Tim Deussen, XRBB) frames the hard problem as “proof”: regulation asks a production to account for itself at the level of the finished work, tools mark at the level of a single output, and authorship/rights live at the level of the individual asset - nothing today carries a claim across all three, so people carry it instead.

AI Agent Pipeline (ARC + Generation)

The hybrid production pipeline of One Small Crawl: authentic archival footage retrieved by agents, AI-generated character/VFX layers, and human compositing. Agents operate under the Agent Production Bible with approval tokens and audit trails.

ARC pipeline (Scenes 1-4 backgrounds)

AGT-016 ARC Retrieval -> searches archives for era-authentic footage
AGT-019 Enhancement   -> upscales, stabilizes, inpaints modern elements
AGT-003 Gen-Video     -> generates Snail character animation layer
AGT-011 Color Pipeline-> era-matches color between archive + AI layers
COMPOSITE (human VFX artist in Nuke/[resolve](../entities/tools/resolve.md))

Full generation pipeline

  • AGT-002/003 Gen-Image/Video from prompts + style refs (Snail character, sc03 color transition, sc05).

Alpha-channel protocol

  • Snail generated on transparent alpha (EXR RGBA, A = character matte), NOT baked into the archive background. Layers:
    1. Archive Background (asset_origin archive or ai_enhanced)
    2. Character Animation (ai_generated, with alpha)
    3. VFX Elements (ai_generated, with alpha)
    4. Color Grade (AGT-011 era LUT/CDL)
    5. COMPOSITE (ai_composite) by human in Nuke/resolve
  • Why: provenance clarity, creative control, legal defensibility (show the unmodified archive plate), iteration speed (regenerate only Layer 2).

asset_origin

Every asset tagged: archive / ai_enhanced / ai_generated / ai_composite. Critical for rights management and IBC Accelerator transparency. Encoded in the filename field 5 (naming-schema).

“Ship of Theseus” rule

Enhancement ratio tracked - see bible 6.3 - so archive authenticity is never silently exceeded.

Notes

AI-Assisted Post (topaz / enhancement)

The post pipeline uses AI enhancement and generation to transform archive material into the film’s on-screen look. The authoritative bible formalizes this as the hybrid pipeline: archive retrieval + AI enhancement + AI generation + human compositing (ai-agent-pipeline).

Evidence on disk

  • 121 full-res TIFFs in Scratch/accelerator/topaz/ - topaz Video AI upscale/enhance output (ai-generated-frames)
  • 94 proxy clips named chf3_prob4_hyp1 - generative clip family with a parameter suffix (prob4/hyp1 = internal pipeline naming)
  • IBC_AI clips 4.x - demo/experiment pieces for the IBC 2026 AI-in-media showcase (ibc-ai-clips)
  • CedricTests in ce’s project - experimental AI passes by Cedric Lejeune (cedric-lejeune; alias “ce” on the collaboration tree, see team)

Authoritative framing

  • AGT-019 Enhancement Agent performs upscales, stabilization, and inpainting of modern elements; AGT-011 Color Pipeline era-matches archive + AI layers.
  • Enhancement ratio (“Ship of Theseus” rule, bible 6.3) tracks how much of an archive frame may be altered.
  • Every enhanced asset carries asset_origin: ai_enhanced (naming-schema).

Notes

AI Copyright

The legal right to claim creative ownership over a particular asset or piece of media, whether produced with AI or not. A central tension in the FRAMES project was between achievable quality and ensuring copyright claimability.

Definition (from the paper)

  • Copyright is the legal right to claim creative ownership over an asset or piece of media, whether produced with AI or not, specifically for control, reproduction, distribution and display.

Key questions raised

  • How can legitimate ownership and copyright regarding AI-generated content be determined?
  • If content is generated exclusively by AI without sufficient human creative input, can it be legally protected under copyright law, or are competitors free to copy it directly?
  • What rights do original authors retain when their works are included in training sets?
  • How can artists and publishers protect their works when copyrighted assets are extracted without explicit consent, attribution, or compensation?

FRAMES learnings

  • The biggest tension in designing the human-AI hybrid workflow was between achievable quality and ensuring copyright claimability.
  • Some AI tools flag identifiable copyrighted material as inputs; when the production already owns the rights, this safeguard is obstructive, which can lead to choosing less-protective tools for still-legitimate outputs.
  • Legal indemnities (contractual safeguards against copyright infringement lawsuits) vary by training architecture and provider terms; their scope and void conditions are open questions.

AI Disclosure

The action of explaining to external parties how AI has been used on a project. Usually best supported by accurate records of ai-transparency, though it can also be a broader agreed statement (e.g. “AI has been used on this project” / “no AI has been used on this project”).

Definition (from the paper)

  • Disclosure is the action of explaining to external parties how AI has been used on a project.
  • It is usually best supported by accurate transparency records, but can also be a broader agreed statement.

In the FRAMES transparency framework

  • Disclosure happens at different levels, from a short credits statement for audiences and press (Level 1) to detailed disclosure statements for industry and stakeholders (Level 2) and full auditable records for commissioner-level governance and legal review (Level 3). See aimici-transparency-framework.

AI Transparency

Production-level transparency about how and why AI was used on a project, distinct from tool-level transparency about what data a model was trained on. The paper focuses on a record of AI usage and decision-making on the project.

Definition (from the paper)

  • Transparency is usually discussed as a record of what data AI tools were trained on. The paper instead focuses on production-level AI transparency: a record of AI usage and decision-making on the project.
  • To be transparent, a record must be created containing accurate information.

Learnings (FRAMES)

  • Substance: a simple human-readable record of how AI was used is a useful internal reference; it is hard to judge how “significant” AI usage is without digging into specific inputs/outputs.
  • Process: transparency documentation can be started mid-project or even retroactively; decision/tool details can be captured from meeting transcripts and chat logs; information should be reviewed and confirmed by the team before finalisation.

Framework

AIMICI Transparency Framework

A framework designed by aimici and tested on the FRAMES project, focused on bringing transparency and understanding of copyright implications to projects with hybrid AI and human workflows.

Scope of attributes

The framework captures the following scope of attributes:

  • WHAT - details of the project and its key phases
  • WHO - who is responsible for overseeing and reporting on AI usage
  • HOW - the AI use cases and tools used
  • WHEN - when tools were used and what decisions were made on which use cases or tools can proceed or be stopped
  • WHERE - where AI actions took place across a workflow and what assets were affected

Three levels of records

  1. Level 1: Credits Statement - for audiences and press. A short, plain-language statement in the credits (one to three sentences) about the production’s AI use.
  2. Level 2: Production AI card - for industry and stakeholders. A one- to two-page reference detailing AI use cases and tools, risk/visibility/ significance attributes, organised into production stages, with a high-level record of human oversight and the businesses involved. Could be published as a website or hosted document.
  3. Level 3: Auditable AI record - for commissioner-level governance and legal review. Full record including signed declarations, decisions, a high-level actions log, tool and licensing details, records of consent, and granular asset-level data such as machine-readable AI marks (required by the eu-ai-act).

In practice on FRAMES

  • The project produced a reviewed production AI card (Level 2) for an AI-centric production for the first time, and explored how the workflow could feed a version of Level 3 records.
  • The credit statement (Level 1) was still in progress at paper time.
  • The auditable record evolved through the project; machine-readable AI marks emerged on some tools, but the team mostly relied on manual tagging of human vs AI-driven assets in the Streamline platform.

Animatic Workflow

The edit builds scenes from animatics - rough timed previews - before full passes. This matches the authoritative editorial layer: the OMC export includes an OTIO timeline (one_small_crawl_v001) and a 45-shot list, so animatic-first is the proven path from script shots to the final edit.

Evidence on disk (August)

  • Animatic - Scene 5.mp4 (Aug 10) -> later scene5.mp4 (107 MB, Aug 19)
  • Scene4Animatic+Archive.mp4 (Aug 14) - animatic that already layers the archive footage (6403-our-trip-of-usa), then Scene4.mp4 (Aug 17)
  • Scene 2 passes run sc02_0010 -> sc02_0072 + a sc02_v1 (146 MB) - granular versioning of the same scene (scene-cutdowns-august)

Authoritative editorial layer

Why it matters

  • Animatic-first lets the team lock pacing before investing in full AI/archive renders; ties directly to the AI passes of ai-agent-pipeline.
  • Shot ids in animatics (sc02_0010…) match the naming schema (naming-schema).

Notes

Archive Footage Reuse

The film’s core technique: authentic archival footage - retrieved, enhanced, and composited with AI-generated elements - rather than fully synthetic imagery. The hybrid pipeline is explicit in the Agent Production Bible: “This production is NOT fully AI-generated.”

How it shows up

  • Archive layer (Scenes 1-4 backgrounds) comes from the ARC pipeline: AGT-016 retrieves era-authentic footage, AGT-019 enhances (upscale, stabilize, inpaint), AGT-011 era-matches color.
  • RAI archive records are the named archive source (rai-archive-records): Route 66 documentaries (Kilimangiaro, TG2 Si Viaggiare, TG3 1900) and space news (Tg3 Artemis II, Speciale TG1 Moon history).
  • On-disk home-movie “Our Trip of U.S.A. Part Two” (6403-our-trip-of-usa) is the family-road-trip layer - the “then vs now” companion to the RAI material (medium confidence on its exact cut placement).
  • The space thread (artemis-ii) pairs with the road: “one small crawl” vs “one small step”.

Why it matters

  • Provenance: every layer tagged asset_origin; the production can show archive licensors the unmodified plate and the separate AI layers.
  • Legal defensibility and IBC Accelerator transparency requirements.
  • The “Ship of Theseus” enhancement ratio (bible 6.3) guards how much of an archive frame may be altered before it stops being “the archive”.

Notes

C2PA & Content Labelling

C2PA (Coalition for Content Provenance and Authenticity) is the open technical standard for cryptographically binding provenance metadata to content, so that audiences and systems can verify where content came from and whether it was AI-generated or edited. In the FRAMES context it is the primary mechanism for content labelling - marking assets and finished media as pure, modified, or synthetic.

What C2PA provides

  • Content Credentials: tamper-evident provenance attached to media (identity of content, people, devices, software, time, location).
  • Content labelling: audience-facing labels distinguishing pure, modified, and synthetic content.
  • Machine-readable AI marks (relevant to the eu-ai-act).

FRAMES decisions (2026-06-29-frames-call call)

  • Responsible AI was narrowed to transparency, with C2PA content labelling as the mechanism, plus a risk-adjusted framework and decision models for AI tool selection, aligned with the eu-ai-act.
  • Audience labels for pure, modified, and synthetic content were agreed.
  • Technical options discussed: frame-level metadata, server-side certificates, alongside emerging European parallel standards.
  • shamir (eddie-ai rep) to assist with C2PA implementation questions.

In the production

  • eddie-ai plans C2PA compliance (per the 2026-08-17-frames-call call); its prompt conversations are exportable for transparency.
  • The AI Ethics Green Paper (ai-ethics-paper) lists C2PA/CAWG among the existing initiatives (with OpenTimeline, SMPTE RIS-OSVP) that production provenance should rely on where possible.
  • The aimici-transparency-framework Level 3 auditable record can include machine-readable AI marks; in practice the team mostly tagged human vs AI-driven assets by hand in the Streamline platform because C2PA marks were not yet present on all tools in a gatherable form.

Era-Locked Look

The central visual doctrine of One Small Crawl: each of the five scenes emulates a different era of film/video technology, and the era character is applied in POST, never baked into generation. Research-locked in the Era and Visual Look Guide v001 (2026-07-04), validated in resolve era-stack sessions.

The five eras

  • sc01 sf: 1926-window Chicago, 35mm B&W orthochromatic nitrate, 18fps cadence baked at 1.33x, halation strong, coarse grain, soot+smoke as content
  • sc02 snd: 1936-40 Dust Bowl, 35mm pan Plus-X, Academy 1.375, locked 24
  • sc03 kch: July 20 1969, amateur 16mm Kodachrome II (K-12), color bleeds from the rainbow
  • sc04 vhs: mid-1980s consumer VHS color-under, sensor documented (Newvicon vs early CCD); sp/hd sub-codes for the orbital capsule sequence
  • sc05 dig: 2020s full-frame CMOS log, 4K UHD, golden hour

Doctrine (global rules)

  • Generate NEUTRAL plates (era code ntrl) - mono, grain, damage, tint, gates are post-only.
  • Cadence baked at generation (scale keyframe timing), never retimed in comp.
  • Two-level match rule: element -> its neutral backplate (level 1), merged comp -> era donor via scene era stack (level 2). Never match element to donor directly.
  • Grain: transplant real archive scans via Neat Video; synthetic = demo fallback only. Band-limit to era MTF ceiling before grain.
  • Aspect ladder 1.33 -> 1.375 -> 1.37 -> 1.33 -> 1.78, gates as final node.
  • NEVER list: no optical flow on sc01-sc03/archive; no halation/grain on sc04/sc05; no Film Look Creator; don’t correct donor character.

Notes

EU AI Act (2026)

The European Union’s AI regulation, whose requirements came into force during the FRAMES project (2026). It requires productions to account for AI use and, in part, to produce machine-readable AI marks.

Relevance to FRAMES

  • The paper asks “How do we stay in compliance with the EU AI Act of 2026?” and notes that requirements came into force during the project.
  • Machine-readable AI marks (required by the EU AI Act) began appearing on some AI tools during the project, but never on all of them and never in a form the workflow could gather up, so the team tagged human vs AI-driven assets by hand (in the Streamline platform).
  • The aimici-transparency-framework Level 3 auditable record can include machine-readable AI marks required by the Act.

Notes

Asset Naming Schema (OSC v1.0)

The production-wide file naming convention for One Small Crawl, extending TR-VFX-IS with asset origin, agent/vendor codes, era codes, and the OMC version model. Every asset name encodes its full provenance.

Syntax

<showid>_<vfx-sequence>_<vfx-shot>_<image-type>_<origin>_<vendor-code>_<version-code>{_<era>}{_<identifying-description>}{_<spec-version>}{.<frame-number>}.<file-extension>

Example: osc_sc02_0100_anim_gen_a003_v003_nr_snail-walk.1001.exr = show osc, scene 02 shot 0100, animation, AI-generated, agent a003, version v003, era ntrl, snail-walk, frame 1001.

Key fields

  • origin: arc (archive), enh (ai_enhanced), gen (ai_generated), cmp (composite), etc.
  • vendor-code: AGT-xxx agent ids (a003 = Gen-Video, a019 = Enhancement…)
  • era: sf/snd/kch/vhs/dig/ntrl from the era-look-guide
  • version-code: integrates OMC v2.8 versioning
  • omission rule: keep underscores for omitted optional fields

Why it matters

  • asset_origin lets the production answer “what did we do with your footage?” for archive licensors - the legal defensibility of the hybrid pipeline.
  • Era codes keep every layer’s look metadata machine-readable.
  • Naming is the surface of the omc-ontology model; each shot also has an OMC sidecar JSON with the same id.

Notes

OMC Ontology (MovieLabs)

The MovieLabs Ontology for Media Creation (OMC) is the data model that governs how every entity in One Small Crawl is named, classified, related, and exchanged. Schema: OMC-JSON v2.8 (https://movielabs.com/omc/json/schema/v2.8), developed by MovieLabs with DreamWorks, Marvel, Paramount, Sony, Universal, Disney, and Warner Bros.

Five pillars (bible 1.2)

Everything in production is modeled as interactions between five entity types within a Context (Narrative + Production framework):

  • PARTICIPANT (person, department, organization, service) performs TASK
  • TASK (units of work) operates on ASSET and runs on INFRASTRUCTURE
  • ASSET (media files, documents, CG elements) stored in / runs on INFRASTRUCTURE
  • INFRASTRUCTURE (compute, storage, networks, software)

Base entity structure

  • entityType, properties, controlled vocabularies; participants, tasks, assets, context (narrative + production), narrative entities, depictions and portrayals, versions and provenance, compositions, infrastructure, creative work, customData.

How One Small Crawl uses it

  • Every shot has an OMC sidecar JSON: e.g. osc_sc03_0070_comp_gen_a002_v001_kc_rainbow_wide.1001.omc.json (composite, generated by agent a002, v001, era kc).
  • asset_origin is mandatory (archive / ai_generated / ai_enhanced / ai_composite) - provenance, rights, IBC transparency.
  • Naming schema is OMC-version-model integrated (see naming-schema).
  • Script exported as .ncp.json, .jsonld, and .ttl; PeriodO gazetteer as .jsonld and .ttl.

Known gap (from the crosswalk)

OMC v2.8 does not model story time: time:hasTemporalExtent ProperIntervals are empty blank nodes, while capture time IS modeled (omc:ProductionScene, slates, camera metadata). The production fills the narrative side with PeriodO periods (show-ont-periodo-crosswalk).

Notes

Responsible AI

The broad practice of ensuring AI is used within a production in a responsible way. Common themes include copyright, data security, transparency, disclosure, and accountability. In the FRAMES context it is the umbrella concept that the AI Ethics Green Paper (ai-ethics-paper) is built around.

Definition (from the paper)

  • “The broad practice of putting effort towards ensuring that AI is used within a production in a responsible way. Common themes include Copyright, data security etc.”
  • The question is not whether to use AI, but whether one can demonstrate that it is being used responsibly.

Key themes

  • ai-copyright - ownership and claimability of AI-assisted output
  • ai-transparency - production-level record of AI usage and decisions
  • ai-disclosure - explaining AI use to external parties
  • ai-accountability - who is liable when something goes wrong
  • eu-ai-act - regulatory compliance
  • Training and skills (responsible AI training, e.g. ScreenSkills-accredited)

FRAMES practice

  • The project explored responsible AI as a constant thread since the proposal (E&R / T&R track: Ethics & Responsibility, Transparency & Risk).
  • Tradeoffs were necessary between responsible-AI aspects and final quality (e.g. choosing less-protective tools when the production already owns the rights to its material).

Responsible AI Stamp

An industry-standardised, globally-recognised Responsible AI mark proposed by aimici (in development as of the paper, Q4 2026 consultation). It is intended to signify best practice in responsible AI working practice across a range of productions.

Rationale

  • A written credit statement alone still requires legal sign-off and interpretability by audiences who may be at a lower level of AI literacy. Judging whether AI use is “good” or “bad” is not clean cut - the use cases and tools are the deciding factor.
  • On top of a baseline statement, an easily understandable Responsible AI stamp can be added to productions.

Scope

  • Intended to signify best practices across film & TV, scripted and unscripted, and AI-first, hybrid and traditional approaches.
  • Adherence combines appropriately trained staff and transparency documentation (see responsible-ai, aimici-transparency-framework).
  • Further details and consultation open in Q4 2026.

unreal Render Pipeline (discrepancy)

An unreal project “Desert 5.8” rendered an 802-frame PNG sequence (“DesertSequence”) found on disk. EARLIER wiki pages treated this as the film’s desert/moon layer.

Discrepancy with authoritative data

The MovieLabs OMC export (Agent Production Bible v1, era look guide, shot list) describes the pipeline as archive + AI generation + human compositing, with NO unreal renders. The bible lists no unreal step; generation is done by Gen-Image/Gen-Video agents (AGT-002/003); the desert in sc04 is a consumer-VHS era, not a CG render.

Conclusion: the unreal “Desert 5.8” renders are likely an EARLY/parallel experiment NOT part of the authoritative production pipeline. Treat as low-confidence background material until confirmed.

Evidence on disk

  • Project: UnrealEngineProjects/Desert 5.8/
  • Output: Saved/MovieRenders/DesertSequence.0000.png -> .0801.png (802 frames ~= 33s at 24fps)
  • See desert-5-8-renders raw note.

Notes

VES Good Practices

Best-practice guidance drawn from the Visual Effects Society (VES) standards for file naming, crew roles, and color management. Applies to any VFX-flavored production, including the agent-orchestrated pipeline of one-small-crawl. Source of truth for naming/roles is the production’s own naming-schema + agent-production-bible; this page records the VES principles those documents build on.

File naming

Ground: VES Transfer Specifications (vestransferspec.org) plus the compatible MovieLabs / ETC VFX Image Sequence Naming v1.0 (TR-VFX-IS). OSC v1.0 (naming-schema) extends this syntax.

  • Allowed filename chars: A-Z a-z 0-9 - _ . only. Reserved/forbidden: @ # ( ). No whitespace.
  • Fields split by _, sub-fields by -; lowercase used consistently; case never distinguishes two values.
  • <= 255 chars; missing optional fields keep their underscore slot so the field order never shifts.
  • Standard image-sequence syntax: <showid>_<vfx-sequence>_<vfx-shot>_<image-type>_<vendor-code>_<revision-code> with optional -description, +camera-ref, +alternate, .<frame>.<ext>.
  • Fix the “fixed” fields early. Show ID, sequence, shot, and image type cannot change once production starts (they are set by production VFX up front). Revision / description / alternate stay tweakable. Getting the fixed fields wrong is the expensive mistake.
  • Version by revision, never by shot number. New pull or added frames -> bump _v001 to _v002. A different element (fire vs smoke) is a distinct instance, not a version; a same-content variant carries +<alternate> (+4k), and the primary best version omits the tag.
  • Vendor codes are assigned by production VFX only - never self-assigned.
  • Naming is independent of directory naming: a file keeps the same name wherever it lives; never drop filename metadata because “it’s in the folder”.
  • Standard naming lets a VFX pull from editorial deliver to any vendor with no translation and no custom ingest. Filenames are a human convenience - always mirror metadata in the file header / asset DB; never rely on parsing alone.

Role description

Ground: VES On-Set/transfer guidance on the production VFX department, and the VES Handbook role taxonomy. Roles here are VFX-department responsibilities, not the full film crew; see team for the OSC roster.

  • The production-side VFX department is the primary data custodian on a show: supervisor, producer, coordinator, on-set supervisor, data wrangler, VFX editor. They own the fixed naming fields and set vendor codes.
  • VFX supervisor - creative + technical head of the VFX effort. Spans all phases:
    • Pre-production: break the script down scene by scene, identify every VFX shot, estimate shot count / complexity / budget, create technical specs (camera requirements, on-set data-capture protocol, plate guidelines).
    • On set: supervise VFX-heavy shooting, capture lighting references (HDR chrome/gray balls, polarized HDR panoramas), place tracking markers, advise the DP so plates match planned CG.
    • Post: review dailies against approved looks, direct vendor teams through iterations, keep consistency across shots, attend color-timing/grading so VFX and live-action match, sign off deliverables.
  • VFX producer / coordinator - track shot progress, budgets, and vendor deadlines; the glue between creative and production.
  • On-set supervisor / data wrangler - guard the plates: correct camera data, lighting refs, and clean ingest on the day.
  • VFX editor - editorial interface: pulls, breakdowns, temp comps for editorial.

Color management

Ground: VES Handbook of Visual Effects (3rd ed., color management chapter by Joseph Goldstone) and the VES Cinematic Color paper (VES.04). The Handbook’s Three Guidelines:

  1. Make critical color decisions in the screening room, not an artist’s cubicle.
  2. At the desktop, use an open-source or commercial color management system if possible, but do not expect too much of it.
  3. Understand and document the color image encodings produced by the production’s cameras, renderers, film scanners, and the encodings consumed by displays, film recorders, and digital projectors.

The job of any color management system:

  • Interpret image data according to its associated color image encoding.
  • Provide a path to and from a common working space where captured image data can be combined.
  • Display original or combined data as an accurate reproduction of what the final consumer will see.

Other VES/industry color principles:

  • Golden rule of VFX: image regions with no VFX must not be modified. Color conversions applied to photography must be perfectly invertible.
  • Work in a scene-referred (scene-linear) working space; the display / output transform is applied only at the end, never baked into masters. Display transforms are for throwaway deliveries (editorial, previews) only.
  • Do no harm to the plate: delivery to the DI is the highest-fidelity representation of the original photography.
  • Standardize interchange on a well-defined encoding (ACES / OpenColorIO / ASC CDL). Round-trip test: unmodified pixels in a VFX render must match the original plate exactly for the DI.

Applied to One Small Crawl

  • Naming: naming-schema (OSC v1.0) implements the VES/ETC syntax with OSC’s origin, agent/vendor, era, and OMC version fields.
  • Roles: team maps VFX-department duties onto the human + agent crew (AGT-xxx) under the agent-production-bible.
  • Color: era-locked-look applies the VES color discipline to the five-era look - plates generated neutral (era code ntrl), era look applied in post, never baked into generation; matches VES “invertible conversions, no harm to plate.”

Notes

Archive Footage - “Our Trip of U.S.A. Part Two”

Observed asset. A home-movie style archival clip used as the film’s source material.

Technical metadata (observed via ffprobe, mp4 copy)

  • Duration: 890.1 s (~14:50)
  • Size: 92.6 MB (mp4); ~3 MB in filename refers to an old bitrate label (“3mb”)
  • Filename pattern: 6403_HM_Our_Trip_of_U_S_A_Part_Two_01_00_47_21_3mb
    • 6403 - likely film/scan identifier (Kodachrome 6403? unknown - LOW confidence)
    • HM - Home Movie (medium confidence)
    • Our Trip of U.S.A. Part Two - title: a family road-trip movie across the USA
    • 01_00_47_21 - likely timecode or reel marker (LOW confidence)

Copies found on disk

  1. One-Small-Crawl/July/ (m4v) - primary project copy
  2. MediaFiles/Content/ (mp4) - converted copy
  3. One-Small-Crawl/Users/gabri/Desktop/IBC2026/Archivio placeholder/ (m4v) - gabri’s IBC2026 archive placeholder

Significance

  • Backbone of archive-footage-reuse: period home-movie footage of an American road trip, likely paired against the Artemis II space-news archive as a thematic juxtaposition (LOW-MEDIUM confidence).
  • Presence in gabri’s IBC2026 placeholder suggests it was shared for an IBC 2026 context (LOW confidence).

Scene Cutdowns and Animatics (August)

Observed assets. The edit’s scene exports and animatics, dated early-late August 2026.

Files (MediaFiles/August/)

  • sc02_0010.mp4, sc02_0020.mp4, sc02_0030.mp4, sc02_0040.mp4, sc02_0050.mp4, sc02_0051.mp4, sc02_0060.mp4, sc02_0070.mp4, sc02_0071.mp4, sc02_0072.mp4 - scene-02 pass exports, all dated Aug 19
  • sc02_v1.mp4 - 146 MB (Aug 19) - first-version full-scene export (v1 inscription)
  • Scene4.mp4 - 25 MB (Aug 17) - Scene 4 pass
  • scene5.mp4 - 107 MB (Aug 19) - full-quality Scene 5
  • Animatic - Scene 5.mp4 - 34 MB (Aug 10) - animatic pre-render of Scene 5
  • Scene4Animatic+Archive.mp4 -14.7 MB (Aug 14) - animatic Scene 4 layered with archive footage
  • 002_GasStation (2).mp4 -43.6 MB (Aug 17) - gas-station shot (US road-trip icon; “002” = shot/scene number, “(2)” = take or version 2 - LOW confidence)

August proxy files (ce’s project mirror)

  • sc02_0040.mov, sc02_0050.mov, sc02_0051.mov, sc02_0072.mov, scene5.mov

Significance

  • Scenes referenced so far: Scene 2 (sc02), Scene 4, Scene 5 - others likely exist
  • “Animatic + Archive” shows the animatic workflow layering archival footage (see animatic-workflow.
  • sc02_v1 + numbered passes = iterative cutdown/refinement workflow (see ai-assisted-post? no - edit workflow.

AI Ethics, Responsibility, Transparency & Risk for Creative Media Production (FRAMES Green Paper)

IBC RED LOGO HERE

IBC 2026 ACCELERATOR MEDIA INNOVATION PROGRAMME

FRAMES: Federated Retrieval, Agentic Media Environments and Software-Defined Workflows

GREEN PAPER:
An R&D Exploration of AI Ethics, Responsibility, Transparency & Risk for Creative Media Production
Edition One [September 2026]
R&D material for FRAMES Final POC Results

AUTHORS:
IBC Lead/Editor: muki Kulhan, IBC Accelerators
Project Co-Leads & First Authors: Roberto Iacoviello (RAI), Paola Sunna (EBU), Chris Vienneau, Raymond Drury (MovieLabs)

Project Team- First Authors: Tim Deussen (XRBB- Extended Reality Berlin-Brandenburg), Kathryn Webb and Ahsan Mallick (AIMICI), Ben Schofield and Aaron Bhugobaun (GC-SC - The Global Creative & Security Community)

Project Team- Contributing Authors:
John Canning (AMD), Erik Weaver (ETC/USC- Entertainment Technology Center at University of Southern California), Florian Reimann(MovieLabs), Clive Santamaria (ITV), John Maxwell Hobbs (Streamline Media), Nick Kennedy (BBC), Nicolas Frei (Enigmatic), Shamir Alibhai (Press Play Labs)

TABLE of CONTENTS

PART ONE: 2

Executive Summary 2

Introduction 3

Context and Scope 3

Key Project Drivers 5

PART TWO: Project Learnings & Insights 6

Learnings 6

How do we stay in Compliance with the EU AI Act of 2026? 7

PART THREE: Creating best practice for applied Responsible AI Governance & transparency 9

Responsible AI best practice 9

Training 9

Transparency 9

The framework for AI transparency on productions 10

Framework in practice on FRAMES 11

In development / coming soon: An industry-standardised globally-recognised Responsible AI stamp 11

PART FOUR: 12

REFERENCES 13

APPENDIX 14

PART ONE:

Executive Summary

AI is already an integral part of production workflows in the media world, spanning everything from script development and screenwriting to production and post-production. However, the rapid adoption of AI brings with it complex ethical, legal, and governance challenges. The question is not whether to use it, but whether one can demonstrate that it is being used responsibly.

This paper, stemming from the work of the IBC Accelerator FRAMES, offers an overview of AI-driven content and media experiences, with a focus on ethics and accountability. It identifies the most pressing challenges and issues, providing insights and practical guidelines to help stakeholders implement AI responsibly within the media sector. Neglecting these aspects could put an organisation on the wrong side of a very expensive conversation.

Introduction

Media industries are leveraging AI to transform key stages of the media value chain, ranging from production to distribution and monetisation.

AI is being integrated throughout the entire pipeline, from scriptwriting and screenwriting to music composition, editing, and animation creation. Much of the AI skills and experience is concentrated in a small group of freelancers working on the frontier, AI skills are in short supply

However, these advancements also raise critical questions regarding authenticity, attribution, copyright, transparency, accountability, and ethical use These all present a new set of commercial risks that need to be quantified to drive executive focus, the mitigation will depend on the potential costs especially when considering use of IP and archive materials. It is becoming clear that these models and platforms present a new set of security risks that require a different approach to current controls and governance.

The situation is further complicated by industry dynamics, organisational structures, and the skill sets of human professionals.

Global regulatory frameworks differ significantly, requiring companies to adapt responsibly to regional requirements.

To embed responsibility into AI systems within the media sector, monitoring and evaluation efforts are guided by several key concerns and priorities: copyright, transparency, privacy and data protection, and evolving policies and regulations. This necessitates the development of shared industry best practices and ongoing stakeholder engagement.

Context and Scope

FRAMES is one of the eight projects selected for the IBC Accelerator 2026 program.

The acronym stands for “Federated Retrieval, Agentic Media Environments, and Software Defined Workflows.”

The project involves a team comprising major media industry organisations and vendors, including RAI, EBU, and MovieLabs as co-leads, alongside members such as XRBB, ETC-USC, ITV, Enigmatic, Yamdu, Eddie AI, Streamline, GC-SC, and AIMICI, and is supported by AMD, Google Cloud, and Shure.

Its goal is to create an end-to-end framework for media production based on AI agents and a structure provided by the “MovieLabs Ontology for Media Creation.” The project’s scope encompasses AI-assisted media production, focusing on pre-production and production phases, and involves key activities such as scriptwriting, character generation, shot and image creation, and background music composition, all essential for producing a final video of professional quality.

While the market offers numerous AI tools and agents, AI cannot become a genuine competitive advantage for an organisation without a robust framework covering data, skills, governance, and accountability. The scenarios analysed within the project provided a meaningful context for examining the challenges and considerations involved in defining a framework for responsible AI in the media sector.

Definitions

Within this paper, we will refer to the following terms - this is what they mean in the context of media production when AI is involved

Copyright - The legal right to claim creative ownership over a particular asset or piece of media, whether produced with AI or not. Specifically for the purposes of control, reproduction, distribution and display.

Transparency - Typically, transparency is discussed in the context of a record of what kind of data AI tools have been trained on. However, in this paper we are primarily focused on the transparency around AI usage and decision-making on the project - production-level AI transparency, not tool-level. To be transparent about this, a record must be created containing accurate information.

Disclosure - This is the action of explaining to external parties how AI has been used on a project. This is usually best supported by accurate records of transparency, although it can also be a broader agreed statement (eg. AI has been used on this project, no AI has been used on this project etc.)

Responsible AI - The broad practice of putting effort towards ensuring that AI is used within a production in a responsible way. Common themes include Copyright, data security etc.

Soloprenuer definition

EU AI Act

Key Project Drivers

The adoption of AI in the media production sector is driven by the need for innovation, productivity improvements, and enhanced human capabilities, whilst keeping creative integrity intact. With this in mind, and within their short five-month sprint, this IBC Accelerator team prioritised routinely exploring these aspects, while also asking questions about AI Ethics, Responsibility, Risk, and Transparency that remain fundamentally important in media production. These questions were explored and left for further exploration through an extension of this research and paper. These aspects include, but are not limited to, questions such as:

  • How can legitimate ownership and copyright regarding AI-generated content be determined?
  • If content is generated exclusively by artificial intelligence without sufficient human creative input, can it be legally protected under copyright law, or are competitors free to copy it directly?

Who is liable for due diligence and risk assessment, especially if something goes wrong? The prompt engineer? The production company?The technical team? The AI model provider or the creator platform? More importantly, at what stage of the production should this be considered?

What exactly are the guarantees of the legal indemnities? These are contractual safeguards designed to protect users against copyright infringement lawsuits arising from generative AI output or training data. However, the scope and applicability of such protections vary significantly depending on the training architecture and the provider’s terms of service. Under what conditions do such indemnities become completely void?

What rights do original authors retain when their works are included in training sets?

How can artists and publishers protect their works when billions of copyrighted assets are extracted without explicit consent, attribution, or financial compensation?

PART TWO: Project Learnings & Insights

Learnings

Copyright

  • The biggest overall tension when designing the best human-AI hybrid workflow on the project was between achievable quality and ensuring copyright claimability.
  • The only way to truly know the copyright-ability of a final piece of media and its content (eg. characters)

Transparency

  • Substance
    • Having a simple human-readable record explaining how AI has been used on the project is really helpful as an internal project reference, helping to build awareness and understanding of how and why tools have been considered within the process.
    • It’s difficult to make judgements on how “important” or “significant” AI usage is for any particular use case without digging further into specific usage actions (inputs, outputs etc.).
  • Process
    • Although it is best to establish a process for transparency documentation from the start of a project, it is possible to achieve it mid-way through or even retroactively.
    • On projects where communication happens mostly online,some decision-making and tool usage details can be captured from meeting transcripts and chat conversations, rather than relying solely on proactive self-reporting or deep technical documentation. Using AI tools responsibly to support this effort is also possible. However ultimately these approaches are still error-prone and information should always be reviewed and confirmed by the team before finalisation.
    • Even if information can’t be documented in real time from start to end of a project, it is important to at least have a responsible person or people who know the project and process well enough to help capture the info retroactively with confidence

Other

  • Tradeoffs were necessary between different responsible AI aspects of the project in order to achieve acceptable quality of the final film, in line with professional-grade animation standards. For example:
    • Some AI tools have safeguards in place for flagging the use of identifiable copyrighted material as inputs, however, when you already own the rights to the content this kind of functionality is obstructive rather than useful. This can lead to choosing tools that are less protective to achieve (still legitimate) outputs

How do we stay in Compliance with the EU AI Act of 2026? Industry insights:

While larger studios continue to integrate AI technologies into their operations, a growing number of independent, small-to-midsize, and AI-native studios are emerging. These entities frequently operate independently of broader industry initiatives, executing productions with minimal centralized studio oversight while rapidly navigating evolving AI governance standards. Consequently, any ethical best practices and framework developments must account for sole proprietors and AI-native studios amid ongoing industry disruption. Furthermore, there is a clear need for a global industry initiative to establish an approved directory of AI tools, evaluating vendors on operational practices, commercial viability, ethical track record, and transparency.

PART THREE: Creating best practice for applied Responsible AI Governance & transparency

Responsible AI best practice

[Summary including reference to IBC last years paper, other responsible ai approaches]

AIMICI has been a partner to the project and also has been working on standards around responsible AI best practice in industry over the last two years. In their approach, they have identified two key levers to help the industry navigate AI concerns at scale:

Training

Not to be confused with AI training, this educational effort is about ensuring everyone working on productions, especially key decision-makers, have an understanding of what AI can do, it’s common risks, a baseline understanding of how AI governance and escalation processes work, and the role each of us play within it.
This is applied training that includes,

By ensuring this kind of training is included, not just skills or tools training, it builds a safer foundation for innovation as tools continue to progress and improve.

This has been established in industry so far via standardised responsible AI training courses that have been accredited by UK skills training body ScreenSkills, released in May 2026 and rolled out through organisations such as Pact.

Transparency

With the foundation of training in place, the issue of lack of transparency of AI usage across productions can now also be tackled fully. Productions must be able to accurately document and appropriately disclose the use of AI on their productions, internally and externally.

The framework for AI transparency on productions

A proposed new framework has been designed by AIMICI and tested on the project, focused on how to bring transparency and understanding of copyright implications to projects that have hybrid AI and human workflows such as this.

The transparency framework, in summary, looks to capture the following scope of attributes

  • WHAT - Details of the project and it’s key phases
  • WHO - Who is responsible for overseeing and reporting on AI usage across those stages
  • HOW - What are the AI use cases and tools used
  • WHEN - When were tools used and what decisions were made on which use cases or tools can proceed or be stopped
  • WHERE - Where did AI actions take place across a workflow, and what assets were affected

The key to applied transparency is ensuring appropriate documentation and records are in place, so that appropriate disclosure is possible at different levels. The AIMICI framework proposes the following three levels of records:

Level 1: Credits Statement - For audiences and press A short, plain-language statement in the credits (one to three individual sentences or statements) that the production creates according to the uses’ importance to the production and/or its compliance needs.
Level 2: Production AI card - For industry & stakeholders One-to-two-page reference detailing AI use cases and tools, and relevant attributes around risk, visibility and significance of the AI use. This is organised into production stages, and includes a high-level record of human oversight over the governance process, and the businesses involved in the record-keeping across the production. Also includes a more detailed set of disclosure statements that form the “long list” from which the Level 1 individual statements are chosen. This could be made publicly available as a website or hosted document.
Level 3: Auditable AI record - For commissioner-level governance & legal review Full record including signed declarations, decisions and high level actions log and further details about tools and licensing at a minimum. Can also include records of consent, and granular actions & asset-level data such as the machine-readable AI marks, which are required by the EU AI act.

Framework in practice on FRAMES

Within the FRAMES project, there was an opportunity to create a production AI card / level 2 record for an AI-centric production process for the first time. We also explored how the team’s proposed workflow could feed into a version of Level 3 records. We explored these two approaches in tandem.

Outcomes:

  • We were able to produce a reviewed production AI card
  • Credit statement is still in progress (this is finalised at the end of the project based on finalise records)
  • The auditable record has evolved throughout the project. With the EU AI act requirements coming into force during the project, we began to see some machine-readable AI marks emerge on some of the AI tools that were used. However ultimately we had to rely mostly on manual tagging of human vs AI-driven assets and workflows within the Streamline platform

In development / coming soon: An industry-standardised globally-recognised Responsible AI stamp

Agreeing a standardised format for a credit statement / on-screen disclosure is a must-do for the industry. However, a written statement alone still requires both legal sign off and intepretability by audiences who may still be at a lower level of AI literacy. For example, judging whether it is “good” or “bad” that a production has used AI to support VFX activities is not clean cut - the use cases and tools are the deciding factor on how responsibly this has been done.

Therefore, on top of a baseline statement, a further evolution has also been proposed by AIMICI - an easily understandable Responsible AI stamp that can be added to productions.

There have been attempts at this in a number of ways already across industry, (REFERENCES) however there has not yet been anything that considers responsible AI working practice at large on a production.

By adhering to the combined requirements of appropriately trained staff and transparency documentation, AIMICI’s new Responsible AI mark is intended to signify best practices across a range of productions including film & TV, scripted & unscripted, and AI-first, hybrid and traditional approaches.

Further details and consultation opportunities will open in Q4 2026. Learn more here:

PART FOUR:

Findings from the project in the context of the AI EU act

The “soloprenuer” in content production, actual best practice on the frontier for major studios, advertisers and content creators

There are a set of emerging standards that could help drive best practice in the use of AI in production. The rate of innovation in models, commercial pressures on traditional media distribution and advertising and the rapid inflows of investment capital have made this technology extremely fast moving. The baseline set of basic metadata including the identity of content, people, devices and software together with time and location that need to be captured to support proper attribution and audit of new workflows. Where possible this should rely on existing initiatives such as OpenTimeline, SMPTE RIS-OSVP, C2PA/CAWG. The experience gained through this project suggests some summary best practice guides that could help build new AI skills for current creative talent.
A closing word from Tim Deussen, XRBB
Five months of building FRAMES taught us that the hard problem in AI-assisted production is not generation. It is proof.
Almost every difficulty we met had the same shape. We could push for the quality a professional animation demands, or we could keep a clean claim to copyright, and holding both at once meant slowing down and writing things down. Machine-readable AI marks began appearing on our tools while the project was still running, but never on all of them, and never in a form our workflow could gather up, so in the end we tagged human and AI-driven work by hand. Safeguards designed to protect an anonymous user got in the way of a production that already owned the rights to its own material. And none of the record could be handed to software: what was made, by whom, and why was written down by people who knew the work.
That is not a complaint about the technology. It is a description of a missing layer. Regulation asks a production to account for itself at the level of the finished work. Our tools mark at the level of a single output. Authorship and rights live at the level of the individual asset. Nothing today carries a claim across all three, so people carry it instead.
None of us can close that gap alone, and a market of incompatible answers helps no one, least of all the artists whose work is at stake. So this paper ends with an invitation rather than a conclusion. We ask broadcasters, studios, producers, vendors, standards bodies and creator organisations to work with us on common standards for ethical AI, production transparency and artistic provenance: a disclosure record that holds good in more than one jurisdiction, provenance that survives post-production, and attribution that reaches the asset and its author, not only the finished master. Build it once, together, and international co-production stays possible. Leave it, and every territory will answer differently, and the people who pay for the difference will be the ones who made the work.

REFERENCES

  • List of AI MODELS T&Cs/Policies (?) - to help support ethics / what to keep eyes peeled for with that company / know BEFORE you go! / BEFORE you start to create (?) / what platforms are doing what / changing what (as of Aug/Sept.2026)
  • eg: HIGGSFIELD - loads of top creators/producers won’t work with them now, because they treated their creators like crap and did not value their worth
  • HOW DO ROLL DOUCHEBAG FREE, when all we want to do is create and make great entertainment, etc?
  • Production Identity Whitepaper
  • EIDR.org content identity
  • EBU - standards?
  • AIMICI-
  • MOVIELABS etc
  • ETC -
  • AMD/HP
  • GC-SCBest practice (Ben/Aaron)
    • Measurement
    • Review & continuous improvement
  • Why do this - quantifiable benefits, creative friction
  • Skills development pathway

APPENDIX

AI-Generated Frames - topaz + prob4_hyp1 Clips

Observed assets. Two families of generated/AI-enhanced imagery with a consistent prob4_hyp1 naming convention.

1. topaz TIFF sequence (121 frames)

  • Path: Scratch/accelerator/[topaz](../../entities/tools/topaz.md)/
  • Pattern: eed46847157f4a6aa3ab92812e683be4_prob4_hyp1_24_000000.tiff …
  • 121 TIFF files, ~6.7 MB each, dated Jul 5 12:14
  • _24_ suggests 24 fps (MEDIUM confidence); eed46847... is likely a source-media hash
  • Folder “topaz” -> output of topaz Video AI (upscale/enhance) (MEDIUM confidence)

2. July proxy clips (94 files)

  • Path: July/Proxy/
  • Pattern: 1 (2)_chf3_prob4_hyp1_000000.mov …
  • 94 proxy clips, same prob4_hyp1 suffix
  • chf3 / prob4 / hyp1 likely encode generation parameters: chunk/format=3, probability=4, hypothesis=1 (LOW confidence - internal pipeline naming)

Significance

  • The film’s post pipeline uses AI generation/enhancement on the archive material (see ai-assisted-post).
  • TIFFs = full-res intermediate frames; .mov proxies = the same content at preview resolution.
  • Scratch/accelerator + topaz folders suggest an experimental enhancement/acceleration workflow (MEDIUM confidence).

IBC_AI Clips 4.1-4.5

Observed assets. A set of short proxy clips under the IBC_AI folder in gabri’s RAI OneDrive.

Files

  • 4.1.mov - 5.2 MB (Aug 31)
  • 4.2.mov - 6.8 MB (Aug 31)
  • 4.3.mov - 9.6 MB (Aug 14)
  • 4.5.mov - 6.7 MB (Aug 31)

Significance

  • Folder name IBC_AI + gabri’s Desktop/IBC2026 folder -> likely demo/short content prepared for IBC 2026 (International Broadcasting Convention) around AI in media (LOW-MEDIUM confidence).
  • The 4.x naming suggests a numbered sequence of clips (a reel or scene breakdown) (LOW confidence).
  • Related experiment pipeline: ai-assisted-post.

Agent Production Bible v1 (IBC2026-SNAIL-001)

Authoritative system context for all AI agents on One Small Crawl. Classified AGENT CONTEXT; agents MUST treat constraints as hard boundaries.

Document control

  • production_id: IBC2026-SNAIL-001
  • bible_type: agent, bible_version 1, effective 2026-05-16
  • maintained_by: AI Supervisor
  • working_title: One Small Crawl
  • context: IBC Accelerator 2026 - animated short film

Key content

  • Section 1: MovieLabs OMC (Ontology for Media Creation) foundation. OMC-JSON v2.8 schema (https://movielabs.com/omc/json/schema/v2.8). Five pillars, base entity structure, participants, tasks, assets, context, narrative entities, depictions, versions, compositions, infrastructure, creative work, customData.
  • Section 2: Production metadata - identity (yaml) + style DNA (see below).
  • Section 3: Hard constraints.
  • Section 4: File system, access control, staging write protocol, file naming convention (cross-refs naming schema).
  • Section 5: Provenance and logging, consent check protocol.
  • Section 6: Agent registry summary, ARC pipeline, compositing alpha-channel protocol, “Ship of Theseus” enhancement ratio, software versioning.
  • Section 7: Task specification schema + worked examples (POST-002 color transition, sc01 silent animation, sc04 Nutella zero-g, POST-004 sound design, RETRIEVE-001 archive retrieval, ENHANCE-001 upscale/inpaint, ENHANCE-002 generative extension).
  • Section 8: Orchestration protocol, inter-agent messages, error handling.
  • Section 9: Approval pipeline with approval tokens.
  • Section 10: SAG-AFTRA and union compliance context.
  • Section 11: Quick reference card.

Production identity (2.1)

  • format: animated_short
  • production_method: hybrid - archive_retrieval + ai_enhancement + ai_generation
  • archive_sources: RAI Archive + [additional archive partners]
  • archive_eras: 1920s silent, 1950s-60s newsreel, 1969 broadcast, 1970s-2000s mixed
  • ai_generation_scope: character animation (Snail), environment extension, color transition VFX, sound design, era-style enhancement
  • characters: CHAR-001 Snail (protagonist, walks Route 66 Chicago to Santa Monica across decades), CHAR-002 Voice From TV (NASA “the Eagle has landed”)
  • scene_count: 5
  • runtime_target: 5-8 minutes
  • delivery_platform: IBC Accelerator 2026 showcase; festival circuit
  • product_placement: Nutella jar floating zero-g in lunar orbit capsule (sc04), clearance PENDING

Style DNA (2.2)

  • time-evolving visual+audio style: each scene emulates a different era
  • character design: cartoon snail, round eyes, simple, expressive, consistent across eras
  • aspect evolution: sc01 4:3, sc02 4:3, sc03 4:3 to wider with color, sc04 4:3 VHS to 16:9, sc05 16:9 or 2.39:1
  • resolution_target: 4K UHD; color_space: ACES AP1 working, Rec2020 delivery
  • sc01: 1920s silent cinema, heavy flicker, vertical scratches, coarse nitrate grain, variable 16-18fps feel, Model T, Chaplin/Keaton refs
  • sc02: 1950s-60s newsreel, 16mm warm rich grain, Cadillacs/Woodies/trucks/Harley

Notes

  • This document WINS on governance/safety/approval; era look guide wins on look parameters.
  • Agents MUST tag every asset with asset_origin (archive vs ai_generated vs ai_enhanced vs ai_composite) - critical for provenance, rights, IBC transparency.
  • See omc-ontology, ai-agent-pipeline, era-locked-look, one-small-crawl.

OSC Agent Registry (OMC JSON)

Registry of AI agents registered for the One Small Crawl production, exported in OMC JSON form. Agents referenced by AGT-xxx ids in the bible and naming schema.

Registered agent roles (from registry + bible)

  • AGT-020 Shot-List Agent (produced the shot list 2026-06-07)
  • Script-Analysis Agent: coverage, structure analysis, characters
  • Gen-Image Agent: text-to-image, image-to-image, inpainting, outpainting, style transfer (AGT-002 in pipeline diagrams)
  • Gen-Video Agent: text-to-video, image-to-video, video-to-video for animated sequences (AGT-003)
  • Gen-Audio Agent: TTS, voice cloning (SAG-AFTRA consent), SFX generation (AGT-004)
  • ASR / Transcription Agent: speech recognition, diarization, timestamping
  • Translation Agent: translation, subtitle localization, terminology
  • AI Metadata Agent: face detection (consent required), object recognition, shot-type classification
  • QC Agent: format compliance, audio levels, sync, subtitle timing, color space
  • Scheduling Agent: schedule optimization, conflict detection, day-out-of-days
  • Report-Gen Agent: call sheets, production reports, cost reports
  • ARC Retrieval Agent (AGT-016): searches archives for era-authentic footage
  • Enhancement Agent (AGT-019): upscales, stabilizes, inpaints modern elements
  • Color Pipeline Agent (AGT-011): era-matches color between archive + AI layers
  • Orchestrator Agent: coordinates, fallback exists (bible 8.7)

Notes

OSC Asset Naming Schema v1.0

Production-specific extension of TR-VFX-IS naming conventions, integrating the OMC version model and the new Asset Origin + Era Code fields.

General syntax (3.1)

  • Allowed: alphanumeric, 0-9, -, _
  • No whitespace; parameters separated by underscore _; dashes - replace whitespace within a parameter
  • Lowercase preferred, case used consistently
  • Under 255 characters

Field sequence (3.2)

<showid>_<vfx-sequence>_<vfx-shot>_<image-type>_<origin>_<vendor-code>_<version-code>{_<era>}{_<identifying-description>}{_<spec-version>}{.<frame-number>}.<file-extension>

Positions: 1 showid, 2 vfx-sequence, 3 vfx-shot, 4 image-type, 5 origin (NEW, HC-011), 6 vendor-code (extended for agents), 7 version-code (+ OMC v2.8 versioning), 8 era (optional, NEW - era lock), 9 identifying-description, 10 spec-version, 11 frame-number (single-frame only), 12 extension.

Omission rule: if an optional field is omitted, surrounding underscores must still be present. Example: osc_sc03_0100_comp_cmp_a003_v003___sv01.00.1001.exr

Example names (3.3)

  • Archive-retrieved (unmodified): osc_sc02_0100_ap_arc_rai_v001_nr_rt66-diner-ext.mov
  • AI-enhanced archive (4K upscale + inpaint): osc_sc02_0100_ep_enh_a019_d001_nr_rt66-diner-ext-upscaled-inpainted.mov
  • AI-generated character animation (with alpha, per-frame): osc_sc02_0100_anim_gen_a003_v003_nr_snail-walk.1001.exr
  • Color transition VFX (sc03 signature): osc_sc03_0400_ctrn_gen_a003_v005_kc_rainbow-bleed.mov
  • Final composite (archive BG + AI FG + VFX): osc_sc02_0100_FINAL_cmp_hum_v001_nr.mov
  • Sound design element: osc_sc03_0000_sfx_gen_a004_v002__rain-on-asphalt.wav
  • Archive temporal extension: osc_sc04_0300_extd_enh_a019_d001_vh_rt66-decay-extended.mov

Asset origin codes (2.5)

  • arc: archive (retrieved, unmodified)
  • enh: ai_enhanced
  • gen: ai_generated
  • cmp: composite
  • (plus others per schema)

Notes

Dashboards and Look Review (OMC export)

Interactive dashboards and review documents shipped in the MovieLabs OMC export. These are the production’s analysis/QC surfaces.

Files

  • osc_camera_dashboard (1).html - camera/shot-size dashboard
  • osc_narrative_vs_complexity.html - narrative vs shot complexity
  • osc_shot_complexity_breakdown (1).html - per-shot complexity breakdown
  • OSC-Look-Review.pdf - era look review deck (companion to era-look-guide)

Notes

Era and Visual Look Guide v001

Per-scene narrative, camera, stock and look specification. Companion to the Agent Production Bible. WINS over the Bible on look parameters; Bible wins on governance. Research-locked 2026-07-04, built/validated in resolve era-stack sessions 2026-07-04/05.

Global doctrine

  • sc01 cadence is 18 fps (speed factor 1.33x), NOT 16 fps (1.5x) - cached recipes stale.
  • archive_is_immovable: synthetic matches archive, never the reverse.
  • era_look_is_post_only: generate clean NEUTRAL plates; era character applied in resolve. neutral_plate_era_code: ntrl.
  • Cadence baked at GENERATION (scale keyframe timing), never retimed in comp.
  • Grain policy: transplant from real archive scans via Neat Video; synthetic grain = demo fallback only.
  • Band-limit before grain: soften to era MTF ceiling BEFORE grain/noise.
  • Two-level match rule (composites): element matched to its own NEUTRAL backplate pre-merge, then merged comp matched to era donor via scene era stack. Elements NEVER match the donor directly.
  • Aspect ladder: 1.33 -> 1.375 -> 1.37 (16mm) -> 1.33 (NTSC) -> 1.78. Gates as final node (osc_all_gate__v###.png). Pillars/letterbox pure black; damage never touches gates.

Era codes (naming field 8)

SceneEra codeStatus
sc01sfratified
sc02sndplaceholder - ratify in naming-key xlsx
sc03kch (color) / snd-adjacent mono (rain)proposed
sc04vhsproposed
sc05digproposed
neutralntrlratified

Scenes

  • sc01 “The Road Begins” (Silent Film, Chicago, 1926-window): 35mm B&W orthochromatic nitrate, Eastman Cine Type F (1201-class). Snail departs from eastern terminus of Route 66 - Michigan Avenue at the Art Institute of Chicago, beneath the lions. Wonder -> determined. Date-dependent set dressing (if 1926: no Buckingham Fountain; Tribune Tower complete). Donor: Wacker 1929 (Newberry). Coal-smoke haze IS content.
  • sc02 “Dust and Motion” (Sound Newsreel, 1936-40, MO-OK-TX): Ariston Cafe (Litchfield, IL, coords 39.1764,-89.6646 verified). Migration years, determined + sad. 35mm pan Plus-X, Academy 1.375, locked 24, 1/48s blur budget.
  • sc03 “The World Turns to Color” (Kodachrome pivot, July 20, 1969, U-Drop Inn, Shamrock TX): amateur 16mm Kodachrome reversal (Kodachrome II, K-12 process, 1961-74). Color bleeds from rainbow. Wonder.
  • sc04 “Somebody’s Tape” (Consumer VHS, mid-1980s, AZ-CA desert): color-under, sensor pick Newvicon vs early CCD, documented. Content. Nutella zero-g capsule shot here.
  • sc05 “Arrival” (Modern digital, 2020s, Santa Monica): full-frame CMOS, log capture, 4K UHD. Content + wonder. Golden hour.
  • sc03/sc04 media transitions are deliberately array-valued in the ontology crosswalk: sc03 opens in sc02 medium, transitions to Kodachrome; sc04 montage drifts Kodachrome -> VHS -> digital.

NEVER list (additive to Bible)

  • NEVER bake era character into generation (plates are ntrl).
  • NEVER retime in comp for cadence.
  • NEVER use optical-flow/frame-blend interpolation on sc01-sc03 or any archive.
  • NEVER apply Halation/Film Grain/film emulation to sc04 or sc05.
  • NEVER use Film Look Creator on any scene.
  • NEVER grade inside damage plugins.
  • NEVER match an element to an era donor directly.
  • NEVER let gates/pillarbox inherit noise, weave, or damage.
  • NEVER “correct” donor character - donor measured values ARE the target.
  • NEVER ship a shot whose OMC sidecar lacks asset_origin, era code, sc03 grade state, lora/dataset provenance (character shots), factual-basis reference (real-place shots).

Notes

OTIO Editorial Timeline (one_small_crawl_v001)

OpenTimelineIO (OTIO) JSON timeline export for One Small Crawl v001. OTIO is the interchange format for edit decisions (EDLs/timelines) used by resolve, Nuke, and other tools.

Content

  • Timeline named one_small_crawl_v001
  • Clips reference the composited shots (sc01-sc05) and archive/AI layers
  • Standard OTIO structure: timeline -> tracks -> clips with source ranges, markers, and metadata (OMC ids where available)

Notes

RAI Archive Records in OMC Export

RAI (Radiotelevisione Italiana) archive records shipped inside the MovieLabs OMC export. These are the archive candidates for the ARC pipeline (ai-agent-pipeline) - the film’s archive layer is RAI material, matching the on-disk RAI OneDrive tree (artemis-ii-tg3-news, ibc-ai-clips).

Records

  • T09143_211 / T26097_303: Tg3 Pomeriggio - Missione spaziale Artemis II (space thread, see artemis-ii)
  • T04331-304: TG3 1900 - C’e una America che vuole salvare la Route 66
  • T08340_222: TG2 Si Viaggiare 2008-2009 - Viaggio lungo la Route 66
  • F157301: Kilimangiaro - Route 66 La strada madre
  • F719859: Kilimangiaro - Route 66 La strada madre (duplicate/related)
  • T12239_123: Speciale TG1 2012-2013 - Storia dell’uomo sulla Luna
  • T94053-151: TG1 Motori 1994 - Harley Davidson
  • T95152-221: Almanacco 1995
  • C10371: Tecnologia - Momenti del progresso
  • C5521: Storia dell’informazione

Notes

  • Route 66 records (T04331, T08340, F157301, F719859) anchor the film’s road narrative in real RAI archival material (route-66).
  • Artemis II + Moon records (T09143, T26097, T12239) feed the space thread.
  • These records are the “archive_sources: RAI Archive” named in the bible.
  • See one-small-crawl, archive-footage-reuse, omc-ontology.

Script v2.2.0 and Shot List (OMC)

Script

  • File: Script-IBC-accelerator-2026-May-16_draft_v2.2.0_20260628T232347Z.ncp.json
  • Draft v2.2.0, updated 2026-06-28T23:23:47Z
  • Story structure: narrative_central with 2 perspectives, 7 dynamics, 4 storypoints, 5 storybeats, 5 moments
  • Logline (from narrative): a snail walks the full length of Route 66 from Chicago to Santa Monica across decades of American media history; each scene is a different era of film/video technology.

Shot list

  • File: shot_list.json
  • Provenance: produced by AGT-020 Shot-List Agent, stage 4.4, provider gemini, model gemini-3.1-pro-preview, created 2026-06-07
  • source_ids include “production/MovieLabs Ontology and Production Data/ Production/08_ARCHIVE/01_lto_manifests/02_metadata/01_raw_ai_enrichment” and creative-vocab-v2.xlsx
  • content_hash: sha256:6954ce2962a99a3ce3b90c040247c11cf918c5ddf94081b6c9e0084295238d7d
  • ontology_version present; 45 shots total
  • Naming: sc01_0010, sc01_0020, …, sc02_…, sc03_…, sc04_…, sc05_… (scene_sequence style)

Shot examples (first shots)

  • sc01_0010 EXT. ROUTE 66, CHICAGO - DAWN | composite | Medium Wide Shot | Worm’s Eye View | snail looks at the Route 66 sign
  • sc01_0030 | Close-Up | Eye Level | sputtering Model T speeds past, ignores snail
  • sc01_0040 | Medium Close-Up | Low Angle | snail makes a quiet, private commitment and begins
  • sc02_0010 EXT. ROUTE 66, MIDWEST - DAY | Medium Shot | Ground Level | traffic roars past on a peak, bustling Route 66
  • sc02_0030 | Close-Up | High Angle | snail pauses outside a diner to listen to music

Notes

Show Ontology x PeriodO Crosswalk v001

Ten PeriodO-conformant period definitions in two authorities: five narrative-era periods (story time Olivia crawls through, one per omc:NarrativeScene) and five capture/playback-medium periods (historical availability window of each capture technology, per the locked Technical Era Spec), joined by oscx:depictedVia and bound to the production graph by UUID.

Audience: MovieLabs / EBU / RAI standards reviewers + osc pipeline.

Key claim

Demonstrates a real OMC v2.8 gap: story time has no OMC class (the graph’s time:hasTemporalExtent ProperIntervals are empty blank nodes), while capture time IS modeled (omc:ProductionScene, slates, camera metadata). PeriodO fills the narrative side; medium periods formalize the production side.

Period examples

  • p-osc-narr-sc1: Route 66 origin era, Chicago, as-scripted 1926 (1926-1927; deny 1927-1929), ortho35
  • p-osc-narr-sc2: Peak Route 66 Midwest era, Ariston Cafe, ~1935-1948 (1935-1938; deny 1946-1948), pan35
  • sc3/sc4 deliberately array-valued: sc03 opens in sc02 medium, transitions to Kodachrome as color bleeds from rainbow; sc04 montage drifts Kodachrome -> VHS -> digital.

Research corpus per period

Verified facts with sources and status, anachronism deny-lists, genAI prompt conditioning, signage generations, hero locations, sound palettes.

Other content

  • oscx:heroLocations: named real locations with NRHP IDs, bypass dates, archival provenance (LOC Highsmith call numbers, Villanova PDTP archive, Burton Frasher/Pomona, etc.)
  • Story Objects & Motifs: me: identifiers (diner, Model T, rainbow, TV, capsule, Nutella, meteor…)
  • NRHP properties DB: Soulsby, Ariston, U-Drop, Blue Swallow
  • Open items: sc03 gas brand (verify via Newberry Curt Teich TX Geographic Index), “End of the Trail” trademark row on p-osc-narr-sc5

Notes

RAI News Archive - Artemis II Space Mission (Tg3)

Observed asset. A proxy clip of an Italian national news report on the Artemis II mission, sourced from RAI’s archives via gabri (RAI).

Technical metadata

  • Proxy file: T26097_303_Tg3_Pomeriggio_Missione_spaziale_Artemis_II.mov
  • T26097_303 - RAI archive identifier (medium confidence)
  • Tg3 Pomeriggio - Tg3 afternoon news programme (high confidence)
  • Subject: Artemis II space mission (NASA crewed lunar flyby)

Location

  • Under gabri’s OneDrive - RAI Radio Televisione Italiana Spa/Archivio/Proxy/
    • the single RAI archive asset found in the project.

Significance

  • Brings the “space” side of the film: news footage of the Artemis II mission, juxtaposed with the vintage American road-trip archive (see archive-footage-reuse).

Original Score - “One Small Crawl Score - Stereo Out”

Observed asset. The film’s original score, exported as a stereo WAV.

Technical metadata (observed via ffprobe)

  • Duration: 270.8 s (~4:31)
  • Size: 52.0 MB
  • Format: .wav, stereo, filename ends “- Stereo Out - 0000” (DAW/resolve export naming)
  • Located in the August folder (both MediaFiles mirror and ce’s project copy).

Significance

  • Confirms the film’s official title: One Small Crawl (score filename).
  • A ~4.5 min score ~= likely the finished film’s duration envelope (LOW-MEDIUM confidence.

Copies found on disk

  1. MediaFiles/One-Small-Crawl/One-Small-Crawl/August/
  2. One-Small-Crawl/Users/ce/Videos/Projects/OneSmallCrawl/One-Small-Crawl/August/

unreal “Desert 5.8” - DesertSequence Renders

Observed asset. A rendered image sequence from an unreal project named “Desert 5.8”.

Technical metadata

  • Engine project: UnrealEngineProjects/Desert 5.8/ (unreal, likely UE 5.8 - MEDIUM confidence)
  • Render output: 802 PNG frames DesertSequence.0000.png -> DesertSequence.0801.png
  • Path: .../Saved/MovieRenders/
  • Content: desert landscape renders (named “DesertSequence”)

Significance

  • Provides the desert/moon-analog imagery layer of the film (see unreal-render-pipeline.
  • with the USA road-trip archive and Artemis II news (archive-footage-reuse.
  • 802 frames ~= 33 seconds at 24fps or ~= 27s at 30fps (computation, not observed.

Wiki Log

Chronological record of all wiki actions. Append-only. Format: ## [YYYY-MM-DD] action | subject Actions: ingest, update, query, lint, create, archive, delete

[2026-09-05] correction | Nicholas Fry -> Nicolas Frei

  • Corrected name throughout wiki: Nicholas Fry was actually Nicolas Frei (Enigmatic, Contributing Author of the AI Ethics Green Paper).
  • Renamed entities/people/nicholas-fry.md -> nicolas-frei.md; updated title + H1.
  • Rewrote all prose + [[wikilinks]] in entities/organizations/enigmatic.md, raw/ethics/ai-ethics-paper.md, log.md, index.md.
  • Rebuilt mdbook + redeployed to https://kitsuli.hollm.cloud/brain/ (verified 200).

[2026-09-02] create | C2PA & content labelling concept page

  • Added concepts/c2pa-content-provenance.md: C2PA (Coalition for Content Provenance and Authenticity) + content labelling - Content Credentials, pure/modified/synthetic audience labels, machine-readable AI marks.
  • Sources: 2026-06-29-frames-call call (C2PA labelling decision), AI Ethics Green Paper (C2PA/CAWG), eddie-ai C2PA compliance, shamir C2PA support role.
  • index.md + SCHEMA.md (c2pa tag) updated.

[2026-09-02] update | snail name resolved to “Stella”

  • Decision found: the FRAMES team agreed the snail is named “Stella” on the 2026-07-13-frames-call weekly call (raw/meetings/2026-07-13-frames-call-frames-call.md), confirmed again on 2026-07-20-frames-call (raw/meetings/2026-07-20-frames-call-frames-call.md).
  • Corrected snail to use “Stella” as the authoritative name (title, H1, new Name section); the PeriodO crosswalk “Olivia” (medium confidence) is now recorded as superseded by the team’s explicit decision, not a live conflict.
  • index.md and meetings.md updated to reflect the resolution (Stella supersedes Olivia).

[2026-09-02] ingest | AI Ethics Green Paper + ethics concepts, AIMICI, cited people

[2026-09-02] update | kitsuli URL /wiki -> /brain (LIVE)

  • Live VPS edit on kitsuli.hollm.cloud (ubuntu@vps-cfec4213, x86_64, /etc/nginx/sites-available/kitsuli.conf), via tools/ssh_run.py + paramiko venv tools/sshbox.
  • Sed-applied (backup at kitsuli.conf.bak.20260902):

    OSC wiki (mdbook static build) served from /var/www/osc-wiki

     -> # OSC brain (mdbook static build) served from /var/www/osc-wiki
    location = /wiki { return 301 /wiki/; } -> location = /brain { return 301 /brain/; } location /wiki/ { alias /var/www/osc-wiki/; … } -> location /brain/ { alias /var/www/osc-wiki/; … }
  • nginx -t OK; systemctl reload nginx OK.
  • Live verification: https://kitsuli.hollm.cloud/brain/ -> 200, title “Index - One Small Crawl Production Brain” https://kitsuli.hollm.cloud/brain -> 301 -> /brain/ https://kitsuli.hollm.cloud/wiki/ -> 200 (Kitsu SPA - no /wiki location exists, falls through to the catch-all location /) https://kitsuli.hollm.cloud/wiki -> 200 (same: Kitsu SPA fallback)
  • python tools/deploy_wiki.py green through step 4 (rebuilt, uploaded, swapped into /var/www/osc-wiki 42M, verify /brain/ 200).
  • No 301 /wiki -> /brain redirect was added per user direction (plain rename; old /wiki links will 404 to Kitsu SPA fallback).

[2026-09-02] update | kitsuli wiki URL -> brain

  • Renamed the OSC mdbook PUBLIC URL path from /wiki to /brain: https://kitsuli.hollm.cloud/brain/ (was https://kitsuli.hollm.cloud/wiki/).
  • Repo-side edits (no filesystem reorg - alias target /var/www/osc-wiki left as-is for now; rename only when convenient):
    • Knowledge/tools/deploy_wiki.py: docstring + verify URL + failure message switched to /brain/.
    • Knowledge/tools/docs/wiki-mcp-deploy.md: table row updated to /brain/ with the caption “static mdbook reader (the OSC Production Brain)”.
    • Knowledge/tools/deploy_wiki_mcp.py: nginx insertion marker comment renamed “# OSC wiki” -> “# OSC brain” so future re-runs pick it up.
  • PENDING on the VPS (out of repo): the live /etc/nginx/sites-available/kitsuli.conf must be edited by hand - rename the location /wiki/ block to location /brain/ (keep alias /var/www/osc-wiki; intact), reload nginx, and optionally add a location = /wiki { return 301 /brain; } redirect so old links still resolve. The deploy script will not detect the mismatch - run python tools/deploy_wiki.py AFTER updating nginx to confirm.

[2026-09-02] update | wikilink pass on shots + production pages

  • Tightened concept references across production entities (50 shot pages, 5 sequence pages, deliverables page, one-small-crawl page).
  • 206 wikilink additions resolved:
    • era-locked-look on 24 shots (era-character phrase “Silent Film Era (1920s)”, “Kodachrome”, “VHS”, “1969 atmospheric”, “2020s full-frame”, etc.)
    • snail on 48 shots (character renamed to Stella in source, page file is still snail.md)
    • kitsu on 55 pages (Kitsu mentions outside existing open in Kitsu markdown links)
    • movielabs on 52 pages (first non-wikilink MovieLabs mention)
    • omc-ontology on 34 pages (first non-wikilink OMC mention)
    • route-66 on 27 pages (first non-wikilink “Route 66” mention)
  • Targets without a wiki page (apollo-11, neat-video, stella) were skipped to avoid introducing broken links.
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | deliverables Master QC added

  • Added a “Master QC” section to deliverables: screening-room color sign-off, per-version (HDR/SDR) display monitoring, display transform never baked in, no-harm-to-plate / round-trip check, technical QC (codec, resolution, frame rate, bit depth, audio), and OMC-logged sign-off - all grounded in the ves-good-practices color discipline.
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | deliverables file naming added

  • Added a “File naming” section to deliverables: deliverable files follow the production-wide naming-schema (OSC v1.0), with VES/ETC good-practice rules (version by revision, fix fixed fields first, filename independent of directory, allowed character set) and an example deliverable stem.
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | deliverables process added

  • Added a “Process” section to deliverables: conform in resolve from the OTIO timeline, sRGB/linear color pipeline for the effect timelines, then finish in ACES (AP1/Rec2020) to deliver both HDR and SDR versions.
  • Rebuilt mdbook reader; verified below.

[2026-09-02] create | deliverables folder in production

  • Added entities/production/deliverables/deliverables.md documenting the film’s output deliverables (final 4K UHD, YouTube export via render-youtube-final-py, score, animatics, OMC sidecars).
  • Added “deliverables” to the build script’s production subfolder list (tools/build_wiki_book.py) so the reader generates a Deliverables cover; added the page to index.md.
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | HOLLM demo wording

  • On hollm, changed “powered by an AMD Strix Halo processor” to “using an AMD Strix Halo processor” in the FRAMES demo description.
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | HOLLM Production Brain wording

  • Removed “(which can be run locally, as in the FRAMES demo, or not)” from the Production Brain description on hollm; now reads “the articulation between an LLM and a semantic wiki based on MovieLabs OMC (omc-ontology) and industry good practices (ves-good-practices)”.
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | HOLLM extends hectools

  • Added to hollm: “Extends: using tools, skills and MCP from www.hectools.com” (after the Hermes Agent basis).
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | HOLLM Production Brain refined

  • Refined the Production Brain description on hollm: it is the articulation between an LLM (runnable locally, as in the FRAMES demo, or not) and a semantic wiki based on MovieLabs OMC (omc-ontology) and industry good practices (ves-good-practices).
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | HOLLM demo hardware + models; tools section

  • HOLLM page (hollm) now documents the IBC Accelerator FRAMES demo stack: HP Z2 mini workstation (hp) with AMD Strix Halo processor (amd), running unsloth inference for the ornith, lfm and qwen models.
  • Created tool pages unsloth, ornith, lfm, qwen; wired hp and amd to the HOLLM demo; added inference/model/unsloth/ornith/ lfm/qwen tags to SCHEMA taxonomy; added all four to index tools section.
  • Aligned render-youtube-final-py summary in index (raw + tools sections).
  • Rebuilt mdbook reader; verified below.

[2026-09-02] update | index.md Tools section added

  • Added “### Entities / tools” section to index.md listing all 22 tool pages with one-line summaries (hollm through render-youtube-final-py).
  • The build script’s tools cover now carries these summaries (rebuilt reader).
  • No page-count change (tools pages already existed).

[2026-09-02] create | HOLLM tool page

  • Added entities/tools/hollm.md: HOLLM, the AI production assistant developed by holli for the mediadata project at insa. Manages the One Small Crawl Production Brain and connects it to the production’s tools. Based on Hermes Agent (by Nous Research); open source.
  • Wired from holli (technology list now links hollm); added holli, insa, open-source tags to SCHEMA.md taxonomy.
  • Tools section is auto-generated by the build script from entities/tools/; rebuilt mdbook reader (HOLLM now in SUMMARY + covers/tools.md).

[2026-09-02] update | Raymond Drewry name correction + email

  • User correction: the AI Ethics Green Paper author’s surname is “Drewry”, not “Drury”. Renamed entities/people/raymond-drury.md -> entities/people/raymond-drewry.md; updated raymond-drewry wikilinks in index.md and log.md.
  • Added email rdrewry@movielabs.com to the person page; noted the paper’s “Raymond Drury” spelling as the same person (raw source kept verbatim).
  • Verified below: ASCII clean, no broken wikilinks, rebuilt mdbook reader.

[2026-09-02] create | VES good practices concept page

  • Added concepts/ves-good-practices.md: VES best-practice guidance on file naming (VES Transfer Spec + MovieLabs/ETC image-sequence syntax), role description (VES Handbook VFX-department taxonomy: supervisor, producer, coordinator, on-set supervisor, data wrangler, VFX editor), and color management (VES Handbook “Three Guidelines” + Cinematic Color).
  • Cross-linked from naming-schema (notes), era-locked-look (notes), and team (notes) pointing to ves-good-practices.
  • Added ves to SCHEMA.md tag taxonomy.
  • Verified below: ASCII clean, no broken wikilinks, rebuilt mdbook reader.

[2026-09-02] update | people enriched from FRAMES meeting notes (12)

  • User request: enrich the people pages from the info collected in the meetings. Added a “## FRAMES weekly calls (2026)” section to 10 person pages (all except the two alias/no-record pages), sourced from the 17 call notes in raw/meetings/.
  • Per page: participant-count on the 17 calls, agreed call role where named (chris-vienneau AI architect; roberto-iacoviello lead AI artist; gabriele-monteleone DP; paola-sunna EBU point person), and dated bullets of key activity/action items with per-call links.
  • Heavy hitters: Iacoviello + Deussen present on all 17 calls (Deussen = Excel shot-list backbone; Iacoviello = showcase/hardware/ rights coordinator, opens the IBC show).
  • john-hobbs active in decisions/actions from 2026-06-08-frames-call (score with suno, Streamline platform) though never in participant tables.
  • aidan-enigmatic + davide-zappia (2 calls each; notes write his name “David”); alberto-ciprian 6; gabriele-monteleone 6; cedric-lejeune 11 (kitsu + resolve infra).
  • aaron-bhugobaun noted as having no call record; the “Super Admin” label (cedric-lejeune) tracked on the person page.
  • Added the meeting tag to 11 pages; verified 0 broken wikilinks and ASCII clean (bytes verified).

[2026-09-02] create | Wiki initialized

  • Domain: One Small Crawl short-film production
  • Structure created: SCHEMA.md, index.md, log.md, raw/, entities/, concepts/, comparisons/, queries/
  • Replaced prior placeholder content (“The Last Light” fictional entry removed).

[2026-09-02] ingest | Production asset survey (batch)

[2026-09-02] ingest | MovieLabs OMC export (11 zips, authoritative production data)

  • Extracted 77 documentary files to Knowledge/movielabs-osc-export/ (bible, era look guide, naming schema, agent registry, script + shot list, OTIO, PeriodO gazetteer, RAI records, dashboards, per-shot OMC sidecars)
  • Raw sources added: agent-production-bible, era-look-guide, asset-naming-schema, agent-registry, script-and-shot-list, otio-timeline, show-ont-periodo-crosswalk, rai-archive-records, dashboards-look-review (raw/movielabs/)
  • Entities rewritten/added: one-small-crawl (authoritative), snail, route-66, team (+agents), artemis-ii (corrected)
  • Concepts rewritten/added: archive-footage-reuse, ai-agent-pipeline, era-locked-look, omc-ontology, naming-schema, animatic-workflow, ai-assisted-post; unreal-render-pipeline marked contested (discrepancy: UE renders absent from authoritative pipeline)
  • index.md rebuilt (13 layer-2 pages + 17 raw sources); SCHEMA.md taxonomy extended (omc, era-look, agent, ibc2026, naming, route-66, snail, character)
  • ASCII purge + byte-clean verification run (0 non-ASCII bytes)

[2026-09-02] update | Entity taxonomy reorganized

  • Created entities/production/{characters,props,environments}/ and entities/team/organizations/
  • Moved snail to entities/production/characters/, route-66 to entities/production/environments/
  • Wikilinks unaffected (basename resolution); index.md sections updated
  • props/ and organizations/ left empty (Nutella prop, RAI org are candidates)

[2026-09-02] update | Organizations added (web research)

  • Populated entities/team/organizations/ with 6 orgs: rai, ebu, movielabs, streamline-media, holli, insa
  • Context: all six are in the IBC2026 Accelerator FRAMES consortium (“Federated Retrieval, Agentic Media Environment and Software Defined Workflows”) - proposed by RAI/EBU/MovieLabs; INSA also Champion; Streamline Media Participant; HOLLI supporting Participant (per IBC2026 press release and show.ibc.org project page)
  • MovieLabs identified as OMC author (bible s1.1); RAI archive source; EBU/INSA consortium governance; HOLLI sustainability tie-in (SC-008); Streamline Media entity identity flagged LOW confidence/UNRESOLVED
  • SCHEMA.md: tag taxonomy extended (organization); index.md updated (now 19 layer-2 pages)

[2026-09-02] update | INSA campus confirmed

  • Production team confirmed the FRAMES participant is INSA Hauts-de-France, Valenciennes (Campus Mont-Houy), not INSA Rennes
  • insa updated: campus UNRESOLVED flag removed; confidence medium to high; page now notes INSA Hauts-de-France audiovisual/multimedia programme
  • index.md line updated to “INSA Hauts-de-France (Valenciennes), FRAMES Champion”

[2026-09-02] update | kitsu links moved into Shot data section

  • Shot pages: kitsu link moved from Notes into “## Shot data (kitsu)” section, first bullet, as clickable markdown link “[open in kitsu](url)”.
  • Sequence pages: existing kitsu link in Facts made clickable the same way.
  • 55/55 pages; verified 0 non-ASCII bytes, 0 broken wikilinks.

[2026-09-02] update | kitsu links on sequence + shot pages

  • Added “- kitsu: http://localhost:5000/productions/b16f9216-…/<shots|sequences>/” to all 55 pages (5 sequences + 50 shots) in entities/production/.
  • URL format verified 200 via curl; links use each entity’s kitsu UUID from the local instance (project One Small Crawl).
  • Verified 0 non-ASCII bytes; 0 broken wikilinks.

[2026-09-02] update | shots enriched with MovieLabs export JSON

  • Enriched all 50 shot pages (entities/production/shots/) with a “MovieLabs export (JSON)” section merged from movielabs-osc-export:
    • generation briefs v0.3.0 (001__generation_briefs_v0.3.0_r02.json) - 50/50
    • shot list v0.2.0 revised (001__shot_list_v0.2.0_comment_v1_r06_revised.json) with AGT-020 shot_list.json fallback - 46+44 codes
    • per-shot OMC sidecar (001__osc_*.omc.json) - 34/50: model, seed, era, provider, task_id, deliverable stem, governor agents, reproducibility
  • Notes on conflicts: kitsu kind vs export kind differ on some shots (sc02_0040/0060, sc03_0050, sc04_0090-0140, sc05_0010/0030-0050); export records kept verbatim alongside kitsu data.
  • index.md + one-small-crawl.md updated with enrichment note; verified 0 non-ASCII bytes and 0 broken wikilinks after the pass.

[2026-09-02] create | sequences + shots pages from local kitsu

  • Added folders entities/production/sequences/ and entities/production/shots/
  • Pulled from local kitsu (localhost:5000, project One Small Crawl): 5 sequences (sc01-sc05, code pages) and 50 shot records (scXX_NNNN)
  • One page per sequence (5) and per shot (50); shot pages carry full kitsu payload: kind (archive/synthetic/composite), reference_archive, beat, beat_summary, beat_shift, tempo, transition, camera, timing, archive_id, timecodes, continuity_flag, canceled flag, preview file id
  • 18 of the 50 shots are canceled in kitsu (sc03_0060-0100, sc04_0100-0220 subset); kept as pages with [canceled] markers
  • index.md: added sequences + shots sections; total pages 20 -> 87
  • source of truth: kitsu-localhost (authoritative); script-and-shot-list raw (45 shots) predates the kitsu breakdown of 50

[2026-09-02] update | people/ flattened to entities root

  • Moved entities/team/team.md into entities/team/people/, then moved the whole people/ folder up to entities/people/ (team/ dir removed)
  • team.md now lives at entities/people/team.md alongside the 12 person pages
  • No wikilink changes needed (links resolve by basename)
  • index.md section renamed to “Entities / people”

[2026-09-02] create | per-person pages for team/people/

  • Created entities/team/people/ with one md per kitsu account on One Small Crawl (12): chris-vienneau, davide-zappia, paola-sunna, cedric-lejeune, roberto-iacoviello, gabriele-monteleone, tim-deussen, john-hobbs, aidan-enigmatic, alberto-ciprian, super-admin, aaron-bhugobaun
  • team.md Humans list and index.md Entities/team now link to each person page
  • super-admin is documented as an alias of cedric-lejeune (same person, two kitsu accounts)

[2026-09-02] update | team collected from local kitsu

  • Queried local kitsu (localhost:5000, logged in as cedric.lejeune@holli.st) for the One Small Crawl production (project b16f9216-7ada-4325-811e-07d900e40f77)
  • Authoritative team: 12 kitsu accounts registered on the production (RAI: Davide Zappia, Roberto Iacoviello, Gabriele Monteleone, alberto ciprian; EBU: Paola Sunna; HOLLI: Cedric Lejeune + Super Admin; FRAMES partners: John Hobbs (Streamline-Media), Aidan Enigmatic, Aaron Bhugobaun (Chapter Forge Media), Chris Vienneau, Tim Deussen)
  • team.md Humans section updated: kitsu list now authoritative, on-disk tree kept as secondary (confidence high)

[2026-09-02] update | frames moved to organizations/

  • Moved frames from entities/ root to entities/organizations/frames.md (user request)
  • index.md: frames listed under “### Entities / organizations”

[2026-09-02] update | team folder + team.md moved

  • Created entities/team/ and moved team into it (team/team.md)
  • index.md: team listed under “### Entities / team”

[2026-09-02] update | archives folder + artemis-ii moved

  • Created entities/production/archives/ and moved artemis-ii into it (user request; space-news thread grouped with production archives)
  • index.md: artemis-ii listed under “### Entities / production/archives”

[2026-09-02] update | one-small-crawl moved to production/

  • Moved one-small-crawl to entities/production/one-small-crawl.md (user request; the film page belongs with the other production entities: characters/snail, environments/route-66)
  • index.md: film listed under “### Entities / production”

[2026-09-02] update | Organizations moved to entities/organizations/

  • User correction: previous flattening to entities/ root was not wanted. Moved rai, ebu, movielabs, streamline-media, holli, insa into entities/organizations/
  • index.md organizations listed under “### Entities / organizations”

[2026-09-02] create | FRAMES project overview

  • New entity frames: overall description of the IBC2026 Accelerator umbrella project “FRAMES: Federated Retrieval, Agentic Media Environment and Software Defined Workflows” that hosts One Small Crawl as its showcase POC (source: show.ibc.org FRAMES project page + IBC2026 press release)
  • Covers objectives, core challenge, approach (OMC script breakdown, archive tagging, local agent swarm, event-driven collaboration), consortium (champions/participants) and relationship to one-small-crawl
  • index.md updated (now 20 layer-2 pages)

[2026-09-02] update | INSA full name corrected

  • Production team states the official full name as “Institut National de Sciences Appliquees” (previously written “des Sciences Appliquees”)
  • insa Facts updated; accents folded to ASCII per strict ASCII policy

[2026-09-02] ingest | FRAMES weekly call meeting notes (17 docx)

  • Source: “Weekly Calls FRAMES.zip” (attachments, 17 AI-generated meeting notes, 30.03.26 to 17.08.26). docx body lives in word/document2.xml; extracted via script to text before ingest.
  • Created raw/meetings/ with one source page per call (17 total), each with participants, key decisions, follow-ups and rich wikilinks to people/orgs/concepts/entities.
  • Created meetings hub (type: summary) listing all calls newest-first with headline decisions + a chronological “what changed when” timeline and cross-cutting decisions log (work streams, tools adopted, ethics/ provenance, archive reality).
  • index.md: new “## Meetings” section (hub + 17 calls); raw-sources count noted; SCHEMA.md tag taxonomy extended (meeting, marketing, audio, metadata, compliance, transparency, kitsu, google-cloud, ai-editor, hardware, color, ethics, frames).

[2026-09-02] update | snail name conflict flagged (Stella vs Olivia)

  • Meeting notes 2026-07-13-frames-call and 2026-07-20-frames-call record the team agreeing the snail’s name is “Stella” (title “One Small Crawl” also confirmed).
  • snail authoritative PeriodO crosswalk uses “Olivia” (medium confidence). Both positions now noted with dates/sources per update policy; flagged for user review.
  • 2026-07-13-frames-call source follow-up writes “One Small Crow” - treated as transcription error for “One Small Crawl”.

[2026-09-02] update | meeting pages charted (17)

  • Rebuilt all 17 raw/meetings/*.md pages into a consistent charted template: At-a-glance table, participants table (name/org-role), Key decisions table (#, Area, Decision, Follow-up ref), Action items table (#, Action, Owner)+ Notes + rich wikilinks.
  • Series numbers (#1-#17), entry filenames and all decisions/actions preserved from the original notes.

[2026-09-02] update | hub pointers tweaked

  • meetings.md hub and index.md unchanged in scope; each call page now links back via the hub section (per-call blurbs already present).

[2026-09-02] update | meeting pages enriched with rich links (17)

  • User request (screenshot of 2026-08-03-frames-call decisions table): enrich meeting pages with inline wikilinks to existing pages.
  • Added across all 17 raw/meetings/*.md: scene refs -> sc01..sc05 (e.g. “Scene 3 (sc03)”), people -> their person pages (Ciprian, Tim, Chris, Cedric, Paola, Roberto, Monteleone, Aidan, John Hobbs, Zappia/David), concepts/tools -> animatic-workflow, unreal-render-pipeline, omc-ontology, streamline-media, and canonical wraps for snail / route-66.
  • 397 wikilink instances across the 17 pages; zero broken links; ASCII clean. Fixed a pre-existing typo (“train thee” -> “train the”).

[2026-09-02] update | entity pages augmented with people + roles

  • User request: add people mentioned in entities and analyse each person’s role to augment the entity MDs.
  • Added per-org people sections with role analysis:
    • rai.md: “Crew on One Small Crawl (RAI)” - roberto-iacoviello (lead AI artist/coordinator), gabriele-monteleone (DP; alias “gabri”), alberto-ciprian (artist/lora/compositing), davide-zappia (archive metadata); replaced the bare “gabri” bullet.
    • holli.md: founders line now links cedric-lejeune; “People” section (cedric-lejeune + super-admin alias; Franck Isabel noted as non-crew).
    • ebu.md: “People” section - paola-sunna (EBU-side point person).
    • streamline-media.md: “People” section - john-hobbs (score + platform), flagged as in-production corroboration of the unresolved identity.
    • ai-assisted-post.md: CedricTests line links cedric-lejeune (“ce” alias).
    • snail.md: compositing bullet links alberto-ciprian as workflow owner.
  • Verified: zero broken wikilinks in entities+concepts; ASCII clean.

[2026-09-02] update | Cedric Lejeune external role (mediadata @ INSA)

  • User-reported fact (2026-09-02): Cedric Lejeune works as tech manager for the mediadata project at insa (outside the One Small Crawl production).
  • cedric-lejeune Notes updated; sources tag extended with user-report.
  • Distinct from his HOLLI/kitsu on the production (see the person page).
  • CLARIFIED same day: mediadata is a partnership between holli and insa; Cedric represents it in FRAMES as its tech manager. Person-page note reframed accordingly.

[2026-09-02] update | Super Admin is a role, not a person (correction)

  • User correction: “Super Admin” does NOT exist as a separate person/account; it is the role/position label for Cedric Lejeune on kitsu (instance admin).
  • Deleted entities/people/super-admin.md. Cedric’s page now states “Super Admin” is his kitsu label (not a separate account).
  • Removed the super-admin alias/link from team.md (team is now 11 kitsu accounts), index.md people list, and holli.md (alias ref dropped).
  • Note: historical log entries below still describe the old “alias” framing (kept as a changelog record).

[2026-09-02] update | pages created for meeting-cited entities (5) + links

Created entities/organizations/{amd,hp,eddie-ai}.md and entities/people/{muki,shamir}.md (AMD, HP, eddie-ai FRAMES Participants; muki IBC coordinator; Shamir eddie-ai rep; ADI = eddie-ai alias). Linked citations in 16 raw/meetings pages (57 new link instances), meetings.md hub, frames.md, streamline-media.md, paola-sunna.md, roberto-iacoviello.md, john-hobbs.md, index.md; added person tag to SCHEMA taxonomy. Link integrity + ASCII clean.

[2026-09-02] update | web URLs made clickable across the wiki

  • Wrapped bare http(s) URLs as markdown links: concepts/omc-ontology.md (OMC-JSON schema), raw/movielabs/agent-production-bible.md (schema URL), log.md (URL list entries) and bare-domain reference entries (show.ibc.org, movielabs.com) across entities/organizations/*.md + log.md prose. kitsu URLs were already [open in kitsu](…) links.
  • log.md template example with <shots|sequences> placeholders intentionally left as plain text (not a real URL).
  • mdbook reader (tools/build_wiki_book.py + OneSmallCrawlBrain-book): added assets/external-links.js and registered it in book.toml additional-js; every external http/https link now opens in a new tab in the built reader (rebuild + verify OK).
  • Entities: confirmed IBC (ibc) org page and muki (muki) IBC-staff wording; John Hobbs company URL (app.streamline-media.net/project/13b69d06-…/work/) recorded on john-hobbs and streamline-media.
  • Verified: no bare http(s) URLs remain outside links (except the log template literal), ASCII clean, 1342 wikilinks with zero broken targets.