video-use - Edit videos through conversation
Transcribes footage and creates conversationally directed edits with cuts, color grading, animations, subtitles, previews, and final renders.
Tags
Updated: 2026-10-16Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Transcribe video footage
- Pack phrase-level transcripts
- Propose editing strategies
- Edit on confirmed plans
- Cut at word boundaries
- Apply per-segment grading
- Generate overlay animations
- Compose output-timeline subtitles
- Render previews and finals
- Verify rendered outputs
- Persist project state
Inputs
- Source video files
- Editing preferences
- Target length and aspect ratio
- Aesthetic or brand direction
- Must-preserve moments
- Must-cut moments
- Animation preferences
- Subtitle requirements
- Strategy confirmation
- Natural-language feedback
Outputs
- Packed transcript file
- Edit decision list
- Cached transcript files
- Graded video segments
- Overlay animation assets
- Master subtitle file
- Preview video
- Final video
- Verification artifacts
- Project memory file
Requirements
- ELEVENLABS_API_KEY
- ffmpeg and ffprobe
- Python dependencies
- Node.js and npm for animation slots
- HyperFrames, Remotion, or Manim when used
- Agent tool for parallel animations
- Write access to the video edit directory