Home / MCP / DaVinci Resolve MCP Server

DaVinci Resolve MCP Server

Provides programmatic access to DaVinci Resolve from MCP clients, enabling control and querying through natural language.

python
Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
    "mcpServers": {
        "davinci_resolve_cursor": {
            "command": "./scripts/mcp_resolve-cursor_start",
            "args": [],
            "env": {
                "RESOLVE_SCRIPT_API": "PATH_TO_RESOLVE_API",
                "RESOLVE_SCRIPT_LIB": "PATH_TO_FUSIONSCRIPT_LIB",
                "PYTHONPATH": "..."
            }
        }
    }
}

You can connect AI coding assistants to DaVinci Resolve through a dedicated MCP server that lets you query and control Resolve using natural language. This server runs alongside Resolve, exposing a programmable interface that Cursor or Claude Desktop can use to automate projects, timelines, media, and more.

How to use

Install and start the MCP server, then connect your MCP client (Cursor or Claude Desktop) to the running server. Once connected, you can ask for information about DaVinci Resolve, list projects, open a project, create timelines, add markers, import media, and manage the media pool. The system translates your natural language requests into Resolve actions.

Recommended flow: start Resolve, launch the appropriate MCP client server, verify the client connection, then begin issuing commands such as: get Resolve version, list projects, open a project by name, create a new timeline, or add a marker at the current position. If you want to switch focus to a different page like Edit, Color, or Fusion, request the current page or switch pages as needed.

How to install

Prerequisites you need before installation:

  • Python 3.6+
  • DaVinci Resolve running in the background
  • Optional: Node.js/npm for some features

Step-by-step installation and setup you can follow to get started quickly:

# 1) Create a Python virtual environment
python -m venv venv

# 2) Activate the virtual environment
# macOS/Linux:
source venv/bin/activate
# Windows:
venv\Scripts\activate

# 3) Install dependencies
pip install -r requirements.txt

# 4) Optionally install MCP SDK directly
pip install git+https://github.com/modelcontextprotocol/python-sdk.git

# 5) If needed, set up DaVinci Resolve scripting environment variables (examples shown)
# macOS
export RESOLVE_SCRIPT_API="/Library/Application Support/Blackmagic Design/DaVinci Resolve/Developer/Scripting"
export RESOLVE_SCRIPT_LIB="/Applications/DaVinci Resolve/DaVinci Resolve.app/Contents/Libraries/Fusion/fusionscript.so"
export PYTHONPATH="$PYTHONPATH:$RESOLVE_SCRIPT_API/Modules/"

# Windows
set RESOLVE_SCRIPT_API=C:\ProgramData\Blackmagic Design\DaVinci Resolve\Support\Developer\Scripting
set RESOLVE_SCRIPT_LIB=C:\Program Files\Blackmagic Design\DaVinci Resolve\fusionscript.dll
set PYTHONPATH=%PYTHONPATH%;%RESOLVE_SCRIPT_API%\Modules

# 6) (Optional) Run pre-launch checks or start scripts as described below

Configuration

Configure Cursor or Claude Desktop to connect to the MCP server using the provided start scripts. The server can be started per client or via a universal launcher that can start multiple clients.

Cursor configuration example (macOS) to point Cursor at the MCP server looks like this in your Cursor config file. Use absolute paths for all items.

{
  "mcpServers": {
    "davinci-resolve": {
      "name": "DaVinci Resolve MCP",
      "command": "/path/to/your/venv/bin/python",
      "args": [
        "/path/to/your/davinci-resolve-mcp/src/main.py"
      ]
    }
  }
}

Troubleshooting and tips

Common issues you might encounter and how to address them include ensuring DaVinci Resolve is running before starting the MCP server, confirming environment variables are set, and checking the server logs for errors. If the server cannot connect to Resolve, verify the Resolve scripting paths and Python environment are correctly configured.

Windows and macOS specific notes are included to help you adapt paths and permissions. On Windows, use forward slashes in configuration files and ensure Resolve is running before starting the server. On macOS, ensure scripts have execute permissions and check Console.app for Python-related errors.

Launch options and available launch scripts

After installation you can start the server using client-specific scripts or a universal launcher.

Cursor-specific start (macOS) and Claude Desktop start are provided as dedicated scripts to configure the environment and start the MCP server for each client.

# Cursor (macOS)
chmod +x scripts/mcp_resolve-cursor_start
./scripts/mcp_resolve-cursor_start

# Claude Desktop (macOS)
chmod +x scripts/mcp_resolve-claude_start
./scripts/mcp_resolve-claude_start

Available features

General capabilities include retrieving DaVinci Resolve version and switching current pages (Edit, Color, Fusion, etc.). Project management supports listing projects, getting the current project, opening a project by name, creating new projects, and saving the current project. Timeline operations cover listing timelines, getting current timeline info, creating timelines, switching by name, and adding markers. Media pool operations include listing clips, importing media, creating bins, and adding clips to timelines.

Windows and macOS notes

Windows users should ensure Resolve is installed in the default location and environment variables are set as described. Paths may need adjustment based on your installation. Check logs in the logs/ directory for issues.

macOS users should ensure scripts have execute permissions, verify environment variables, and check Console.app for Python-related errors. DaVinci Resolve must be running before starting the server.

Project structure and development notes

The project provides a main MCP server implementation and a set of helper scripts to install, check readiness, and start the server. The typical workflow uses a quick-start script to set up and run the server in one step.

Notes on future and support

Support is available for issues and feature requests via the issue tracker. The project plans to expand Windows and Linux support, add more DaVinci Resolve features, and extend compatibility with Claude Desktop.

Available tools

Get Resolve version

Query the installed DaVinci Resolve version.

List projects

Return all projects available in Resolve.

Open project

Open a project by its name.

Create project

Create a new DaVinci Resolve project.

Save project

Save the current project.

List timelines

List all timelines in the current project.

Create timeline

Create a new timeline within the current project.

Switch timeline

Switch to a timeline by name.

Add marker

Add a marker at the current timeline position.

List media clips

List clips in the media pool.

Import media

Import a media file into the media pool.

Create media bin

Create a new bin in the media pool.

Add clip to timeline

Add a media clip to the current timeline.