DaVinci MCP Professional

An enterprise-grade Model Context Protocol (MCP) server that exposes the full DaVinci Resolve scripting API to AI assistants. This project is a hard fork of davinci-resolve-mcp by @samuelgursky, rewritten and maintained independently.

Supported MCP clients: Claude Desktop (primary), Gemini CLI, ChatGPT.

---

Prerequisites

• DaVinci Resolve Studio installed and licensed (the free edition does not support external scripting)
• Python 3.10 or later, installed system-wide (the installer from python.org with "Add to PATH" and "Install for all users" selected). On Windows, DaVinci Resolve locates Python through the Windows registry; a uv-managed or user-only Python install will not be found and will cause a crash.
• uv — fast Python package and virtual environment manager

---

Installation

From source (recommended)

git clone https://github.com/Positronikal/davinci-mcp-professional.git
cd davinci-mcp-professional
uv venv
uv sync

Standalone Windows executable

Download the pre-built Windows binaries from Releases. No Python installation required.

The release includes two executables:

Executable	Purpose
davinci-mcp-server.exe	MCP server — launched automatically by Claude Desktop (or another MCP client). Communicates over stdio with no console output. Use this in your claude_desktop_config.json.
davinci-mcp.exe	Interactive CLI — for running the server manually in a terminal. Displays a startup banner, checks that DaVinci Resolve is running, and supports --debug and --skip-checks flags.

Most users only need davinci-mcp-server.exe configured in their MCP client. davinci-mcp.exe is useful for troubleshooting or verifying that the server can connect to Resolve before configuring a client.

---

Configuring Claude Desktop

Locate or create your claude_desktop_config.json file:

• Windows: %APPDATA%\Claude\claude_desktop_config.json
• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

Add the server entry:

Running from source (Windows):

{
  "mcpServers": {
    "davinci-resolve": {
      "name": "DaVinci MCP Professional",
      "command": "C:\\path\\to\\davinci-mcp-professional\\.venv\\Scripts\\python.exe",
      "args": ["C:\\path\\to\\davinci-mcp-professional\\mcp_server.py"]
    }
  }
}

Running from source (macOS):

{
  "mcpServers": {
    "davinci-resolve": {
      "name": "DaVinci MCP Professional",
      "command": "/path/to/davinci-mcp-professional/.venv/bin/python",
      "args": ["/path/to/davinci-mcp-professional/mcp_server.py"]
    }
  }
}

Using the standalone Windows executable:

{
  "mcpServers": {
    "davinci-resolve": {
      "name": "DaVinci MCP Professional",
      "command": "C:\\path\\to\\davinci-mcp-server\\davinci-mcp-server.exe",
      "args": []
    }
  }
}

Restart Claude Desktop after saving the config.

---

Other Supported Clients

Gemini CLI and ChatGPT support the MCP standard and can connect to this server using the same mcp_server.py entry point. Their MCP integration is still maturing — consult each client's documentation for the current configuration method.

---

Basic Usage

1. Start DaVinci Resolve and wait for it to fully load.
2. Start the MCP server (Claude Desktop does this automatically when configured).
3. Ask your AI assistant to interact with Resolve:

What version of DaVinci Resolve is running?
List all projects in the database.
Create a new timeline called "Act 1".
Switch to the Color page.
Import /path/to/clip.mp4 into the media pool.

---

Further Reading

Document	Purpose
USING.md	Developer setup, build instructions, contributing
BUGS.md	Troubleshooting and bug reporting
CONTRIBUTING.md	Contribution guidelines
COPYING.md	GPL-3.0 license

For developer setup, build instructions, and contribution guidelines, see USING.md.