Implements the 2026-04-02 transformation spec (Phases 1-8) and fixes all critical bugs found during 5-pipeline E2E testing. Governance & Decision Intelligence: - Pipeline-specific stage order in checkpoint (replaces global STAGES list) - Provider scoring engine (lib/scoring.py) with 7-dimension weighted ranking - Decision log artifact enforced at proposal/idea stage across all 10 pipelines - Delivery promise classifier prevents silent motion-to-still downgrades - Structured shot language in scene_plan schema (camera, lens, lighting, DOF) - Variation checker and slideshow risk scorer block samey output before render - Creative intake, capability extension, and creative-intake meta skills - Final self-review artifact with 5 mandatory checks before presenting output - Source media review contract for user-supplied footage Render & Theme System: - Remotion AnimatedBackground now derives colors from playbook (no more hardcoded dark blue fintech gradient on every video) - video_compose builds custom ThemeConfig from playbook YAML colors/fonts — custom playbooks flow through to Remotion automatically - Explainer component wires theme to all child components (charts, cards, etc.) - resolveAsset() handles absolute paths on Windows/Unix via file:// URIs - RENDERER_FAMILY_MAP synced with actual Remotion compositions Critical Bug Fixes: - Windows npx subprocess: run_command() resolves .cmd wrappers via shutil.which() - Silent renderer downgrade: Remotion failure now returns explicit error with options instead of silently falling back to FFmpeg - .env inline comment parsing strips trailing # comments from API keys - concat_path UnboundLocalError in video_compose finally block - audio_mixer and showcase_card capture=True kwarg bug - Selector estimate_cost() calls fixed (_select_tool -> _select_best_tool) - asset_manifest schema expanded with provider, license, subtype fields - screen-demo subtitle_gen moved from required to optional tools - Duration drift detection in post-render final review (>25% warns)
2.2 KiB
Script Director - Cinematic Pipeline
When To Use
This stage builds the beat map, selected lines, title-card copy, and reveal structure for the cinematic piece. You are shaping rhythm, not writing a dense explainer.
Prerequisites
| Layer | Resource | Purpose |
|---|---|---|
| Schema | schemas/artifacts/script.schema.json |
Artifact validation |
| Prior artifact | state.artifacts["proposal"]["proposal_packet"] |
Emotional arc and source truth |
| Tools | transcriber, scene_detect |
Optional dialogue mining and source review |
Process
1. Build A Beat Map First
Use a simple structure:
- hook,
- escalation,
- reveal,
- landing.
If the piece is longer, add one midpoint turn. Do not let it become essay-shaped.
2. Use Dialogue Sparingly
If source speech exists, use transcriber to find:
- strong standalone lines,
- emotional phrases,
- concise declarations,
- reveal phrases.
If there is no useful dialogue, keep the script title-led or narration-led and say so in metadata.
3. Keep Title Cards Short
Title-card copy should feel trailer-like:
- fewer words,
- more contrast,
- more whitespace,
- more timing precision.
4. Store Beat Truth In Metadata
Recommended metadata keys:
beat_mapdialogue_selectstitle_card_copymusic_turnssilence_windows
5. Quality Gate
- the beat map escalates cleanly,
- dialogue and title cards do not explain the same thing twice,
- the reveal lands distinctly,
- the landing gives the viewer a final feeling or action.
Mid-Production Fact Verification
If you encounter uncertainty during script writing:
- Use
web_searchto verify factual claims before committing them to the script - Use
web_searchto find reference images for visual accuracy - Log verification in the decision log:
category="visual_accuracy_check"
Every factual claim in the script should be traceable to the research_brief.
If you make a claim that isn't in the research, do additional research and
add the source. Do not invent statistics, dates, or attributions.
Common Pitfalls
- Writing full explanatory paragraphs instead of beats.
- Using too many title cards.
- Revealing the best moment too early.