Wraps FFmpeg and FFprobe for local media processing, providing tools to check dependencies, get media file metadata, extract frames from video as base64 images, extract/convert audio to MP3, and split audio into chunks.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-ffmpeg extract audio from video.mp4 to mp3"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server wrapping FFmpeg and FFprobe for local media processing.
Works with Claude Code, Codex, Claude Desktop, Cursor, VS Code, Windsurf, and any MCP-compatible client.
Node.js 18+
FFmpeg and FFprobe installed and in PATH
Install FFmpeg:
brew install ffmpegRelated MCP server: ffmpeg-mcp
claude mcp add ffmpeg -- npx -y github:pauloFroes/mcp-ffmpegAdd to your codex.toml:
[mcp.ffmpeg]
command = "npx"
args = ["-y", "github:pauloFroes/mcp-ffmpeg"]Add to claude_desktop_config.json:
{
"mcpServers": {
"ffmpeg": {
"command": "npx",
"args": ["-y", "github:pauloFroes/mcp-ffmpeg"]
}
}
}Add to .cursor/mcp.json:
{
"mcpServers": {
"ffmpeg": {
"command": "npx",
"args": ["-y", "github:pauloFroes/mcp-ffmpeg"]
}
}
}Add to .vscode/mcp.json:
{
"servers": {
"ffmpeg": {
"command": "npx",
"args": ["-y", "github:pauloFroes/mcp-ffmpeg"]
}
}
}Add to ~/.windsurf/mcp.json:
{
"mcpServers": {
"ffmpeg": {
"command": "npx",
"args": ["-y", "github:pauloFroes/mcp-ffmpeg"]
}
}
}Tool | Description |
| Check if ffmpeg and ffprobe are installed and available |
| Get media file metadata (duration, resolution, codecs, format) |
| Extract frames from video at regular intervals (returns base64 images) |
| Extract/convert audio from video to MP3 |
| Split audio file into chunks of N minutes |
MIT
Check if ffmpeg and ffprobe are installed and available in PATH. Returns status, version, and installation instructions if missing.
Inputs: no named parameters.
Extract/convert audio from a local video file to MP3. Returns the path to the extracted audio file.
Inputs: file_path (required), output_dir.
Extract frames/screenshots from a local video file at regular intervals. Returns images as base64 that Claude can see and analyze.
Inputs: file_path (required), format, max_frames, interval_seconds.
Get metadata about a local media file (video or audio): duration, resolution, codecs, format, filesize. Uses ffprobe.
Inputs: file_path (required).
Split an audio file into chunks of N minutes. Returns paths to all chunk files. Useful for processing large files that exceed API size limits.
Inputs: file_path (required), output_dir, chunk_minutes.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 5 tools
Each tool has a distinct purpose: checking dependencies, extracting audio, extracting frames, getting media info, and splitting audio. There is no overlap or ambiguity in their functionalities.
All tool names follow a consistent verb_noun pattern in snake_case, such as 'check_dependencies' and 'extract_audio', making them predictable and easy to understand.
With 5 tools, the server is well-scoped for common ffmpeg operations like extraction and metadata retrieval. It is neither too sparse nor too heavy.
The tool set covers essential audio/video extraction and info tasks, but lacks some common ffmpeg features like video conversion or merging. Minor gaps exist but do not severely hinder basic workflows.
Transform video, audio and images, and generate media from prompts. FFmpeg, captions, models.
FFmpeg as a service for AI agents: typed video editing tools, async jobs, downloadable outputs.
Video, audio, and image processing for AI agents: convert, transcribe, upscale - 150+ operations.
Hosted MCP tools for FFmpeg-style video and audio processing through FFMPEG API.
