davinci-resolve

This MCP server exposes a suite of tools to interact with DaVinci Resolve and Fusion through the Model Context Protocol. It supports project management (creating, loading, and saving projects), timeline manipulation (creating timelines, changing the current timeline), media management (importing media, organizing folders, and constructing timelines from clips), and Fusion integration (adding Fusion compositions and creating nodes). Additionally, you can inspect the current project and timeline, navigate Resolve's pages (Edit, Fusion, Color, Fairlight, Deliver), and remotely execute Python or Lua code within Resolve. To use it with Claude Desktop or an MCP client like 5ire, connect to the DaVinci Resolve MCP server and begin issuing the provided commands to perform complex workflows, automate repetitive tasks, or prototype new editing pipelines. Specific commands cover retrieving information (project, timelines, mediapool, status), performing actions (create_project, load_project, save_project, create_timeline, import_media), and running code inside Resolve (execute_python, execute_lua).

Once connected, you can retrieve current project data, examine media pools, and then procedurally build timelines from media, attach Fusion nodes to clips, or render pages and manage storage. The server acts as a bridge between Claude-style assistants and Resolve’s Python API, enabling sophisticated automation and AI-assisted editing workflows directly inside Resolve.

Prerequisites:

• DaVinci Resolve Studio (version 17 or higher recommended)
• Python 3.8 or higher
• Claude Desktop (for AI integration) or another MCP client

Installation steps:

1. Clone the repository and navigate to it: git clone https://github.com/apvlv/davinci-resolve-mcp.git cd davinci-resolve-mcp

2. Install Python dependencies: pip install -r requirements.txt

3. Install the MCP server in Claude Desktop (or your MCP client): mcp install src/resolve_mcp/server.py

   Alternatively, for development work, install with editable mode: mcp install src/resolve_mcp/server.py --with-editable .

4. Run the MCP server locally (example using this package): python -m resolve_mcp.server

Note: If you install via your MCP client, refer to that client's documentation for exact startup commands and connectivity options.

• Ensure DaVinci Resolve is running before connecting with Claude Desktop or your MCP client.
• This MCP server uses Resolve's Python API; Python paths and Resolve installation must be compatible with the server script.
• If you encounter connection issues, check that any required environment variables (such as RESOLVE_PYTHON_PATH) are correctly set to point to the Resolve Python environment.
• When automating, prefer batch operations (e.g., create multiple timelines in sequence) and handle Resolve state changes (loading a project before manipulating timelines).
• The server exposes both information retrieval (Resources) and action-based commands (Project Management, Timeline Management, Media Management, Fusion, and Advanced Operations). Use the appropriate commands for your automation tasks.
• For advanced usage, the execute_python and execute_lua commands let you run arbitrary scripts inside Resolve, which is powerful but should be used with caution to avoid destabilizing the session.