VFX MCP

VFX MCP is an MCP server that provides video processing and editing capabilities to large language models through FFmpeg bindings. Built using FastMCP and ffmpeg-python, the server connects AI assistants directly to local media files and FFmpeg command pipelines. Software developers, content creators, and automation engineers use VFX MCP to perform complex video transformations via natural language or autonomous agent workflows. The server enables LLMs to inspect media metadata, trim clips by timestamps, concatenate multiple files, resize dimensions, and transcode across different video and audio formats. It also supports audio track replacement, custom FFmpeg filter applications, playback speed adjustments, and frame-accurate thumbnail generation. In addition to tool execution, VFX MCP provides resource endpoints for browsing video workspaces and reports real-time progress for long-running media processing jobs.

Category: Design, Media & Creative

Tags: ffmpeg, media manipulation, video editing, video processing

Visit VFX MCP

How to install and configure VFX MCP

1. Ensure Python 3.13+ and FFmpeg are installed on your system. 2. Install the package using pip by running pip install vfx-mcp. 3. Open your Claude Desktop configuration file and add the server to the mcpServers object: json { "mcpServers": { "vfx": { "command": "vfx-mcp", "args": [] } } } 4. Alternatively, if running from a local git clone with uv, configure it with: json { "mcpServers": { "vfx": { "command": "uv", "args": ["run", "python", "/path/to/vfx-mcp/main.py"], "cwd": "/path/to/vfx-mcp" } } } 5. Restart your client to start executing video editing tasks.

What you can do with VFX MCP

• Trimming specific video segments using start timestamps and durations to assemble short clips. - Concatenating multiple video files together with specified transition styles into a continuous montage. - Transcoding video files into web-friendly formats and multiple resolutions with custom video codecs. - Extracting audio streams or replacing original background audio with new music files. - Generating image thumbnails at designated timestamps or extracting frames from video recordings. - Inspecting detailed video metadata including bitrate, resolution, framerate, and codec details.

Key facts

• Open Source
• https://github.com/conneroisu/vfx-mcp
• Design, Media & Creative, Files, Documents & PDFs
• ffmpeg, media manipulation, video editing, video processing

Part of MCP Servers

How do I install VFX MCP?

You can install VFX MCP directly from PyPI using pip install vfx-mcp. Alternatively, clone the repository and run it using the uv package manager with uv sync and uv run python main.py, or enter a Nix development shell using nix develop. Ensure Python 3.13 and FFmpeg are installed.

What can VFX MCP do?

VFX MCP enables an MCP client to perform programmatic video editing tasks. It provides tools for trimming, concatenating videos, converting codecs and formats, changing resolutions, extracting or replacing audio tracks, modifying playback speed, applying FFmpeg filter strings, and extracting frame thumbnails.

Which MCP clients work with VFX MCP?

VFX MCP works with any Model Context Protocol client that supports local command-based server execution, including Claude Desktop, Cursor, or custom Python implementations using the FastMCP Client library.

Is VFX MCP open source?

Yes, VFX MCP is open source software distributed under the MIT License on GitHub. You are free to inspect, modify, and integrate the code into your personal or commercial video processing pipelines.

What system requirements are needed for VFX MCP?

VFX MCP requires Python 3.13 or newer and a system installation of FFmpeg. When using Nix, FFmpeg is installed automatically inside the development shell.