Practical overview
When it fits
Use it when you want local FFmpeg operations exposed to a coding agent or need the same editing tools directly from the command line.
Who it is for
- People using Claude Code, Cursor or Codex who want an agent to carry out video-editing operations.
- Developers using MCP clients or command-line workflows to run local FFmpeg media tools.
Documented uses
- Cut, join or shorten footage by removing silence.
- Reframe video to a target aspect ratio or duration.
- Add captions or karaoke-style highlights using supplied cue text.
- Adjust loudness, clean audio and prepare platform-oriented exports.
Requirements & limits
The public repository is MIT-licensed. The documented setup uses local FFmpeg and Python; the npx installer requires Node.
- FFmpeg 6.1+ and Python 3.10+ are required; Node 16+ is needed only for the npx installer.
- Individual operations depend on FFmpeg codecs and filters, and some builds lack optional components.
- Transcribing speech for captions needs a separately installed local speech engine.
- The toolkit performs editing operations but does not decide what to cut or whether a deliverable is approvable.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install the skill and check local media tools
Set up the skill for a coding agent and check whether the machine has the FFmpeg capabilities its tools require.
Before you start
- FFmpeg 6.1 or later
- Python 3.10 or later
- Node 16 or later to use the npx installer
Steps
Install FFmpeg using the package method appropriate for your operating system; the README gives sudo apt install ffmpeg as an Ubuntu/Debian example.
Run npx ffmpeg-skill to install the skill; use the documented --cursor or --codex option for those agent setups, or --all for all three named agents.
Run npx ffmpeg-skill doctor to check FFmpeg, ffprobe and required components.
Expected outcome
The skill is installed for the selected agent setup, and the doctor command reports available capabilities.
Guide limitations
- Some FFmpeg builds omit filters or libraries needed by particular tools; a missing optional component can make doctor’s overall ok result false while other tools remain usable.
- Speech transcription requires a separately installed local speech engine.
Compare options for AI coding assistance
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Ffmpeg Skill | Cut, join or shorten footage by removing silence. | CLI, Server, Self-hosted | The public repository is MIT-licensed. The documented setup uses local FFmpeg and Python; the npx installer requires Node. |
| Figwright | Convert a selected Figma design into framework-aware code informed by layout, typography, variables and components. | CLI, Server | The project describes the server as free and says the free Figma tier is sufficient. A Figma desktop app is required to import the development plugin. The repository is public and MIT-licensed. |
| Next Devtools MCP | Discover running Next.js development servers and their available runtime tools. | CLI, Server, Self-hosted | The project is published as a public GitHub repository and documents installation through npm tooling. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 3 November 2026
- Original website title
- kajisho5/ffmpeg-skill · Official homepage title · retained in its source language