Vedit-MCP

Vedit-MCP is an MCP server that provides video editing tools to AI agents using standard natural language requests. It connects local video manipulation workflows with an AI client by wrapping the FFmpeg command-line utility. The server is built for video editors, content creators, and automation developers who need to perform programmatic video alterations without manually writing complex FFmpeg command strings. By exposing FFmpeg functions through the Model Context Protocol, Vedit-MCP allows an LLM agent to interpret user intentions, process target footage stored in a local directory, and output modified media. It requires a local installation of FFmpeg and a configured Python runtime to run scripts or serve MCP connections. The system works with sample architectures like google-adk or client tools like Cline, reading source files from a configured workspace folder and writing newly generated video clips to designated output directories automatically based on prompt parameters.

Category: Design, Media & Creative

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

Visit Vedit-MCP

How to install and configure Vedit-MCP

1. Clone the repository or download the source code from https://github.com/zakahan/vedit-mcp. 2. Install the Python dependencies with pip install -r requirements.txt or uv pip install -r requirements.txt. 3. Ensure FFmpeg is installed on your system using brew install ffmpeg on macOS or sudo apt install ffmpeg on Ubuntu. 4. Configure your MCP client (such as Cline in cline_mcp_settings.json) by adding the server configuration: json { "mcpServers": { "vedit-mcp": { "command": "python", "args": [ "vedit_mcp.py", "--kb_dir", "your-kb-dir-here" ] } } } 5. Replace your-kb-dir-here with the directory path containing your source video files.

What you can do with Vedit-MCP

• Trimming and cutting raw video footage into smaller clips by specifying exact start and end timestamps via natural language. * Batch processing localized video clips stored in a knowledge base directory through conversational prompts in an MCP-enabled agent. * Automating media transformations like re-encoding or container switching using underlying FFmpeg integration without manual terminal scripting. * Connecting autonomous agent frameworks such as google-adk to execute programmatic video tasks using reasoning models.

Key facts

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

Part of MCP Servers

What is Vedit-MCP?

Vedit-MCP is an open source Model Context Protocol server that enables AI clients to execute video editing tasks through natural language commands. It works by interpreting model instructions and executing FFmpeg commands directly on video files stored in a designated local directory.

What prerequisites are required to use Vedit-MCP?

You need Python installed along with the packages listed in requirements.txt. Additionally, FFmpeg must be installed on your operating system and accessible from your system PATH, as the script relies on it for media processing.

Which MCP clients work with Vedit-MCP?

The server can be registered in clients supporting local command-based MCP configurations, such as Cline via cline_mcp_settings.json. It can also be integrated into custom agent frameworks like google-adk or standard MCP tool callers that support stdio execution.

Is Vedit-MCP open source?

Yes, Vedit-MCP is an open-source project hosted on GitHub, allowing developers to review the source code, integrate it with their own video pipelines, or build custom editing workflows on top of it.