Choose theme
AI Visuals
ffmpeg-skill
ffmpeg-skill gives a coding agent a set of local tools for editing video and audio with FFmpeg.
Its workflow starts by measuring the source file, uses structured commands for the edit, and checks the output. You can also call the tools directly or through its MCP interface. Use this as a first read, not a recommendation. Open the original project before trusting details like terms, limits, privacy, cost, setup, or safety.
What it is
Local media tools for coding agents
The repository packages an Agent Skill, Python command-line tools, and an MCP server around FFmpeg. Its documented tools cover cuts, joins, captions, audio adjustments, formatting, exports, and output checks.
Why it stands out
Inspect before and after an edit
The skill tells the agent to probe the input before editing and inspect changed video with a contact sheet afterward. Tool contracts expose inputs, outputs, required FFmpeg features, and verification steps.
Availability
A public, local workflow
The project documents installation for Claude Code, Cursor, and Codex. Its Python tools run locally with FFmpeg; Node.js is used for the npx installer. The repository declares an MIT license.
Why it matters
What makes it useful
A coding agent can write an FFmpeg command, but it may assume the wrong frame rate, choose an incompatible format, or finish without checking the file it made. ffmpeg-skill puts source measurements, named editing operations, and output checks in one repeatable workflow. That gives a builder a clearer way to inspect what the agent actually produced.
What to know
Where it fits
Use it when you want an agent to make a concrete edit to local footage: trim an interview, resize a clip, add captions, adjust audio, or prepare an export. The scripts can also be used directly from a terminal. The person or calling agent still chooses the useful footage, cut points, and visual judgment.
Notable points
What stands out
The repository shows an active 2.x release line and describes automated tests on Linux, macOS, and Windows. Its real-device media corpus has been run on Linux and macOS; the project says the full corpus and an independent user installation have not yet been reproduced on Windows. Check the current release and machine-specific results before building a Windows workflow around it.
Before using
What to review
Install Python 3.9 or newer and FFmpeg 5.0 or newer. The npx installer also needs Node.js 16 or newer.
Run the documented doctor command to see which FFmpeg filters and encoders your local build has. Some editing tools need components that a basic FFmpeg package may lack.
Review the agent's planned cut and output. A contact sheet can show visible problems, but selecting the right moment or crop still needs a person or a visual agent.
Allow for re-encoding time and output files when a change cannot use stream copy; keep the original footage available.
Reader fit
Who may find it relevant
Creators and builders who want a coding agent to carry out repeatable local video edits.
People who want to inspect or call the underlying FFmpeg tools without handing footage to a cloud editing service.
Less useful when you need a visual timeline editor or want the software itself to decide which scenes make a good story.
Editorial note
Why LifeHubber lists it
The practical value is the path from a plain-language edit request to measurable local output. Probe the file, make a named change, then check the result. If that is the workflow you need, the repository gives you both the agent instructions and the commands to inspect.
Source links
Source materials
Reader note
Before relying on this entry
LifeHubber lists entries to help readers inspect AI projects, not to endorse them or prove they are safe, suitable, accurate, maintained, or right for a specific use. We do not verify every entry in depth. Before relying on anything listed, review the original materials, terms, privacy practices, limits, and risks that matter for your situation.
More in Productivity
Keep browsing this category
Explore more productivity tools.
Penpot
penpot/penpot
An open-source collaborative design platform with an official MCP server that lets AI agents inspect and modify Penpot files, components, tokens, styles, layouts, and assets through hosted or local connection paths.
OpenPencil
open-pencil/open-pencil
An open-source design editor that opens Figma and Pencil files and gives built-in AI, coding agents, scripts, and custom tools access to the same editable document structure.
Clovy
open-software-network/os-clovy
An open-source Mac AI workspace, formerly named June, for chat, dictation, meeting notes, image generation, and local agent work with project-described approval gates; the project describes app state as local by default and publishes verification evidence for Clovy API.
For project maintainers
Listed here? You can use the badge.
If you maintain a project with a current LifeHubber listing, you may add the optional “Listed on LifeHubber AI Resources” badge to its README, docs, or website. No introduction or permission request is needed.
