#5656th of 82 in Agent frameworks and SDKsFor image, video and design, video and animation
video-use
Agent skill that edits raw video footage into a finished cut via ffmpeg
- Stars
- 30.3k
- License
- MIT
- Last commit
- Oct 2026
- Language
- Python
90 on 10 Apr 202630,303 on 30 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
A skill for Claude Code, Codex and similar shell-capable agents that turns a folder of raw takes into edit/final.mp4. It transcribes each source with ElevenLabs Scribe, packs word-level transcripts into a text file the LLM reads, and renders cuts with ffmpeg. Filler removal, color grading, subtitles and animation overlays are included, followed by a self-evaluation pass on the rendered output.
Who it is for: Creators and engineers editing talking-head or tutorial video via a coding agent
Strengths
- Works from a ~12KB text transcript plus on-demand timeline images instead of dumping frames
- Cuts on word boundaries with 30ms audio fades and silence-gap detection
- Self-eval loop re-checks rendered cut boundaries and re-renders up to 3 times
- Session notes persist in project.md so work resumes across sessions
Weaknesses
- Requires an ElevenLabs API key; transcription is a cloud call
- Needs a separate agent such as Claude Code or Codex; not standalone
- No release published; install is a git clone plus symlink
- Animation overlays depend on extra tools like HyperFrames, Remotion or Manim
What it needs
- no GPU
- Needs ffmpeg, ElevenLabs API (Scribe), Claude Code or Codex, uv or pip, yt-dlp (optional)
- Models: Claude Code, Codex, Hermes, Openclaw
Quick answers
Is video-use open source?
Yes. Its code is on GitHub under the MIT license.
Does video-use need a GPU?
No, it runs without one.
Also in Agent frameworks and SDKs
See all 82| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 95 out of 100 |
| 2 | openai-agents-pythonPython SDK for multi-agent workflows with handoffs, guardrails and tracing | 90 out of 100 |
| 3 | langchainPython framework for building LLM agents and apps from interoperable components | 89 out of 100 |
| #4 | browser-usePython library that lets an LLM agent control a web browser | 89 out of 100 |
| #5 | tradingagentsMulti-agent LLM framework that simulates a trading firm's analysis process | 89 out of 100 |