Ffmpeg_python_mcp

What is Ffmpeg_python_mcp?

Ffmpeg_python_mcp is a Model Context Protocol (MCP) server that integrates FFmpeg with AI assistants. It provides powerful video and audio processing capabilities—format conversion, cutting, merging, effects, and hardware acceleration—designed for developers who want to give their AI tools direct media manipulation abilities.

How to use Ffmpeg_python_mcp?

Ensure Python 3.12+, FFmpeg, and the uv package manager are installed. Clone the repository, run uv sync to install dependencies, then start the server with uv run mcp dev main.py or uv run python main.py. For AI clients like Claude Desktop, add the server configuration to the client’s JSON config file (command: uv run python main.py, working directory: project root, protocol: stdio).

Key features of Ffmpeg_python_mcp

• Full video/audio processing: convert, cut, merge, compress
• Hardware acceleration support (Intel QSV, NVIDIA NVENC)
• Asynchronous concurrent processing for parallel task execution
• Streaming support: M3U8 merging and live stream handling
• Video effects: watermark, GIF conversion, speed change
• AI-friendly interface using the standard MCP protocol

Use cases of Ffmpeg_python_mcp

• Extract audio from a video and convert it to a desired format
• Cut a segment or merge multiple video files into one
• Convert video or audio between formats with codec and quality options
• Add watermarks, resize videos, or create animated GIFs
• Process streaming media by merging M3U8 manifests into MP4 files

FAQ from Ffmpeg_python_mcp

What dependencies are required to use Ffmpeg_python_mcp?

You need Python 3.12 or later, FFmpeg installed on your system (via Homebrew, apt, or the FFmpeg website), and the uv package manager.

How do I configure Ffmpeg_python_mcp with Claude Desktop?

Edit the Claude Desktop configuration file – ~/Library/Application Support/Claude/claude_desktop_config.json on macOS or %APPDATA%\Claude\claude_desktop_config.json on Windows – and add an entry under mcpServers with command uv, args ["run", "python", "/path/to/ffmpeg_python_mcp/main.py"], and the working directory set to the project root.

What hardware acceleration is supported?

The server supports Intel Quick Sync Video (QSV) and NVIDIA NVENC hardware encoding. It includes a check_hardware_acceleration() function and an option to convert video using QSV (convert_video_with_qsv).

What transport protocol does the server use?

The server communicates via the stdio protocol, as shown in the MCP client configuration examples.

Is there a limit on concurrent tasks?

The server supports asynchronous concurrent processing, allowing AI assistants to call multiple tools in parallel, with claimed batch performance improvements of 3–5 times. No explicit task limit is stated in the README.