DaVinci Resolve MCP Server: Give Your AI Assistant Full Control Over Resolve Studio

If you've ever wished you could tell your AI coding assistant "speed up this clip by 20%" or "add a dissolve transition here" without leaving your editor, there's now a tool for that. The DaVinci Resolve MCP Server connects AI agents directly to Resolve Studio, giving them permission to read and manipulate timelines, clips, markers, and project settings through the Model Context Protocol.

This isn't a remote control hack or a limited API wrapper. It's a proper MCP server that exposes Resolve's Python scripting interface to any MCP-compatible client. That means your AI assistant can now act like an actual editing assistant.

What It Does

The server provides a set of tools and resources your AI can use to interact with DaVinci Resolve Studio. Through MCP, it exposes:

• Tools for creating timelines, importing media, adding clips, applying transitions, setting in/out points, and modifying timeline settings
• Resources for reading timeline info, clip metadata, marker positions, and project settings

So instead of manually clicking through Resolve's UI for every small change, you can have an AI model do it programmatically. The server handles the connection to Resolve's Python API, translates MCP requests into API calls, and returns structured data back to your AI.

Why It's Cool

What stands out here is how straightforward it is. The server is built with the official MCP SDK and uses Resolve's existing Python scripting API. No custom daemons, no weird workarounds. You install it, point it at your Resolve instance, and your AI can start editing.

Some practical use cases:

• Automated rough cuts based on markers or script notes
• Batch edits across multiple timeline clips (color grading, speed changes, transitions)
• Scripted assembly from source footage with specific in/out points
• AI-assisted review where the model reads timeline info and suggests changes

The developer took a smart approach by keeping the MCP server as a thin layer over Resolve's native API. That means if you already know how to use Resolve's Python scripting, this feels natural. If you don't, the MCP abstraction makes it simpler to experiment.

How to Try It

You need Python 3.10+ and DaVinci Resolve Studio 18.5 or later installed on the same machine. Resolve's scripting API requires a local connection, so the server and Resolve must be on the same computer.

# Clone the repo
git clone https://github.com/samuelgursky/davinci-resolve-mcp.git
cd davinci-resolve-mcp

# Install dependencies
pip install -r requirements.txt

# Configure your MCP client (Claude Desktop, Cline, etc.)
# Add the server path to your client's MCP config

The README has a detailed setup section for different MCP clients. If you're using Claude Desktop, you'll add the server to claude_desktop_config.json. For Cline, you add it to their MCP settings.

Once connected, you can ask your AI to:

• "Create a new timeline called 'Daily Edit' with the clips in my Bin 1"
• "Add a cross dissolve between clip 2 and clip 3"
• "List all markers on Timeline 1 with their comments"
• "Set timeline resolution to 1920x1080 at 24fps"

Final Thoughts

This is useful if you already work in Resolve and want to automate parts of your edit without writing full Python scripts yourself. The MCP approach lets you treat your AI as a smart editor who can follow instructions and inspect your project state.

Is this going to replace a human editor? No. But if you're a developer who sometimes cuts video, or a VFX artist who wants to batch process timelines, this saves real time. The implementation is clean, the documentation is solid, and it works with the tools you probably already use.

Check it out on GitHub: samuelgursky/davinci-resolve-mcp

Follow us @githubprojects for more developer tools and open source projects.