calesthio
2cd36fa8e0
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)
226 lines
6.6 KiB
YAML
226 lines
6.6 KiB
YAML
name: talking-head
|
||
version: "2.0"
|
||
description: >
|
||
End-to-end talking-head video pipeline. Takes raw footage of a person speaking,
|
||
transcribes it, makes edit decisions, generates subtitles, mixes audio,
|
||
and composes the final output.
|
||
category: talking_head
|
||
stability: beta
|
||
default_checkpoint_policy: guided
|
||
|
||
extensions:
|
||
custom_scripts: true
|
||
custom_playbooks: true
|
||
custom_skills: true
|
||
custom_tools: false
|
||
|
||
required_skills:
|
||
- pipelines/talking-head/executive-producer
|
||
- pipelines/talking-head/idea-director
|
||
- pipelines/talking-head/script-director
|
||
- pipelines/talking-head/scene-director
|
||
- pipelines/talking-head/asset-director
|
||
- pipelines/talking-head/edit-director
|
||
- pipelines/talking-head/compose-director
|
||
- pipelines/talking-head/publish-director
|
||
- meta/reviewer
|
||
- meta/checkpoint-protocol
|
||
|
||
orchestration:
|
||
mode: executive-producer
|
||
skill: pipelines/talking-head/executive-producer
|
||
budget_default_usd: 0.50
|
||
max_revisions_per_stage: 3
|
||
max_send_backs: 3
|
||
|
||
compatible_playbooks:
|
||
- clean-professional
|
||
|
||
stages:
|
||
- name: idea
|
||
skill: pipelines/talking-head/idea-director
|
||
produces:
|
||
- brief
|
||
- decision_log
|
||
tools_available: []
|
||
checkpoint_required: true
|
||
human_approval_default: true
|
||
review_focus:
|
||
- Brief accurately reflects footage content
|
||
- Target platform is appropriate
|
||
- Duration target is realistic
|
||
success_criteria:
|
||
- Schema-valid brief artifact
|
||
- Brief references the provided raw footage
|
||
|
||
- name: script
|
||
skill: pipelines/talking-head/script-director
|
||
required_artifacts_in:
|
||
- brief
|
||
optional_artifacts_in:
|
||
- source_media_review
|
||
produces:
|
||
- script
|
||
required_tools:
|
||
- transcriber # speech-to-text
|
||
optional_tools: []
|
||
tools_available:
|
||
- transcriber
|
||
checkpoint_required: true
|
||
human_approval_default: true
|
||
review_focus:
|
||
- Transcription accuracy (spot-check key sections)
|
||
- Section boundaries align with topic changes
|
||
- Timestamps match actual speech
|
||
success_criteria:
|
||
- Schema-valid script artifact
|
||
- Script sections have valid timestamps from transcript data
|
||
|
||
- name: scene_plan
|
||
skill: pipelines/talking-head/scene-director
|
||
required_artifacts_in:
|
||
- script
|
||
optional_artifacts_in:
|
||
- brief
|
||
produces:
|
||
- scene_plan
|
||
required_tools: []
|
||
optional_tools:
|
||
- frame_sampler # representative frame extraction
|
||
- face_tracker # face position analysis for reframing decisions
|
||
- silence_cutter # silence detection for jump cut planning
|
||
tools_available:
|
||
- frame_sampler
|
||
- face_tracker
|
||
- silence_cutter
|
||
checkpoint_required: true
|
||
human_approval_default: true
|
||
review_focus:
|
||
- Scenes cover full footage duration
|
||
- Enhancement opportunities identified
|
||
- Overlay scenes are feasible
|
||
success_criteria:
|
||
- Schema-valid scene_plan artifact
|
||
- Scenes cover the full duration of the raw footage
|
||
|
||
- name: assets
|
||
skill: pipelines/talking-head/asset-director
|
||
required_artifacts_in:
|
||
- scene_plan
|
||
optional_artifacts_in:
|
||
- script
|
||
produces:
|
||
- asset_manifest
|
||
required_tools:
|
||
- subtitle_gen # captions
|
||
optional_tools:
|
||
- audio_mixer # audio prep
|
||
- image_selector
|
||
tools_available:
|
||
- subtitle_gen
|
||
- audio_mixer
|
||
- image_selector
|
||
checkpoint_required: true
|
||
human_approval_default: false
|
||
review_focus:
|
||
- Subtitle file exists and matches transcript timing
|
||
- Audio extracted and normalized
|
||
- All asset paths resolve to existing files
|
||
success_criteria:
|
||
- Schema-valid asset_manifest artifact
|
||
- All referenced asset files exist on disk
|
||
|
||
- name: edit
|
||
skill: pipelines/talking-head/edit-director
|
||
required_artifacts_in:
|
||
- scene_plan
|
||
- asset_manifest
|
||
optional_artifacts_in:
|
||
- script
|
||
produces:
|
||
- edit_decisions
|
||
optional_tools:
|
||
- silence_cutter # auto jump cuts on silent segments
|
||
- video_trimmer # speed control (0.5x–2x), cut, concat
|
||
tools_available:
|
||
- silence_cutter
|
||
- video_trimmer
|
||
checkpoint_required: true
|
||
human_approval_default: false
|
||
review_focus:
|
||
- Cuts reference valid source files
|
||
- Subtitles enabled with proper styling
|
||
- Audio configuration is complete
|
||
success_criteria:
|
||
- Schema-valid edit_decisions artifact
|
||
- All cuts reference valid source files
|
||
- Timestamps are within source duration bounds
|
||
|
||
- name: compose
|
||
skill: pipelines/talking-head/compose-director
|
||
required_artifacts_in:
|
||
- edit_decisions
|
||
- asset_manifest
|
||
optional_artifacts_in:
|
||
- scene_plan
|
||
produces:
|
||
- render_report
|
||
- final_review
|
||
required_tools:
|
||
- video_compose # rendering
|
||
- audio_mixer # audio layering + segmented music
|
||
optional_tools:
|
||
- face_enhance # face enhancement
|
||
- eye_enhance # eye brightening, dark circle removal
|
||
- color_grade # color correction
|
||
- audio_enhance # audio cleanup
|
||
- auto_reframe # aspect ratio conversion with face tracking
|
||
- face_tracker # face position data for reframing
|
||
- silence_cutter # auto jump cuts
|
||
- video_trimmer # speed control (0.5x–2x)
|
||
- remotion_caption_burn # animated word-by-word captions via Remotion
|
||
- showcase_card # letterboxed showcase cards with typography
|
||
- video_stitch # multi-clip assembly with transitions
|
||
- visual_qa # frame extraction + quality validation
|
||
tools_available:
|
||
- video_compose
|
||
- audio_mixer
|
||
- face_enhance
|
||
- eye_enhance
|
||
- color_grade
|
||
- audio_enhance
|
||
- auto_reframe
|
||
- face_tracker
|
||
- silence_cutter
|
||
- video_trimmer
|
||
- remotion_caption_burn
|
||
- showcase_card
|
||
- video_stitch
|
||
- visual_qa
|
||
checkpoint_required: true
|
||
human_approval_default: false
|
||
review_focus:
|
||
- Output video exists and is playable
|
||
- Subtitles are synced with speech
|
||
- Audio is clear with balanced levels
|
||
success_criteria:
|
||
- Schema-valid render_report artifact
|
||
- Output video file exists and is playable
|
||
|
||
- name: publish
|
||
skill: pipelines/talking-head/publish-director
|
||
required_artifacts_in:
|
||
- render_report
|
||
- final_review
|
||
produces:
|
||
- publish_log
|
||
tools_available: []
|
||
checkpoint_required: true
|
||
human_approval_default: true
|
||
review_focus:
|
||
- Metadata is complete and compelling
|
||
- Export package is structured correctly
|
||
success_criteria:
|
||
- Schema-valid publish_log artifact
|
||
- Export package or publish URLs are present
|