zroeschen/Video-Production-Factory

Edit a video

Keep each project in video-projects/<slug>/, preserving the source recording. Read the workspace guide and the project's brief or DESIGN.md. If a direction is missing, propose a style from style-library/registry.json based on the footage.

1. Inspect duration, streams, and source resolution with ffprobe. Copy the source into the project's assets or reference an explicitly provided local path.
2. Read docs/TOOLS-AND-API-KEYS.md before service setup. Reuse a matching word transcript. Otherwise use the student's chosen provider; Zander defaults to node scripts/transcribe-elevenlabs.mjs <source>. OpenAI Whisper and local Whisper are supported workflow alternatives after transcript normalization. Check available credentials or local dependencies. Explain uploads and costs before any service call whose authorization is still missing.
3. Read ../cut-silences/SKILL.md. Produce an EDL and a retimed transcript, and render the silence pass when editing is authorized.
4. Read ../cut-mistakes/SKILL.md. Inspect candidates in context and preserve intentional emphasis. Record the reviewed cuts. If no mistakes need cutting, carry forward the silence output. Match every transcript to its actual video.
5. Read ../video-storytelling/SKILL.md for the visual arc and ../hyperframes-video-beats/SKILL.md for overlays. Write a beat sheet with transcript anchors, one visual idea per beat, and deliberate callbacks.
6. Read MOTION_PHILOSOPHY.md, ../hyperframes/SKILL.md, and relevant GSAP references before authoring. Use ../style-library/SKILL.md to select or adapt a card. Localize every asset used by the composition.
7. Run preflight and HyperFrames lint from the project, plus the transcript-sync validator for anchored sub-compositions. Review Studio before the draft.
8. Render a draft, inspect encoded frames and transitions, and listen across cut boundaries. Check face framing, text legibility, black frames, and A/V sync. Resolve failures before the final render. Follow any review approvals already provided; do not repeat approval requests for the same authorized action.

Deliver the final MP4 path, edit decisions, retimed transcript, composition, and VERIFY.md describing what was checked and any remaining limitations. An automated detector proposes edits; editorial judgment remains necessary.