# samuelgursky/davinci-resolve-mcp [Health: Active]

**Category:** 🎨 Art & Culture  
**Repository:** https://github.com/samuelgursky/davinci-resolve-mcp  
**GitHub Stars:** 2666  
**npm Downloads (last month):** 27093  
**Views:** 5  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/samuelgursky-davinci-resolve-mcp

## Description
MCP server integration for DaVinci Resolve providing powerful tools for video editing, color grading, media management, and project control

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "davinci-resolve-mcp": {
    "command": "npx",
    "args": ["-y","davinci-resolve-mcp"]
  }
}
```

## Documentation

## What samuelgursky/davinci-resolve-mcp MCP server does

The samuelgursky/davinci-resolve-mcp MCP server gives MCP-compatible AI clients access to DaVinci Resolve operations. Its live-control path uses Resolve's official Scripting API and covers editing, media pool organization, render configuration, review markers, color grading, Fusion, Fairlight, and project lifecycle operations. The repository also describes extension authoring and source-safe media analysis capabilities.

A local browser control panel is included for inspecting Resolve state, running source-safe analysis, reviewing analyzed clips and shots, and editing analysis output. The package additionally ships an optional Node-based advanced server for working with `.drp`, `.drt`, and `.drx` files while Resolve is not running. That offline server is intended for file and database/XML-level changes rather than live application control.

## How it works

The default compound mode exposes 36 MCP tools. Related Resolve actions are grouped behind action parameters to reduce context usage. Full mode exposes 353 more granular tools, with one MCP tool for each Resolve API method; it can be started with `src/server.py --full` or through `src/resolve_mcp_server.py`.

For DaVinci Resolve Studio, external scripting must be enabled under Preferences > General > External scripting using > Local. The free edition cannot accept ordinary external scripting, so the project provides an in-application bridge. That bridge runs from Resolve's Workspace > Scripts menu and re-exports the live Resolve object through an authenticated loopback listener. The server uses the bridge automatically when external scripting is unavailable.

The advanced server works without Resolve running and can operate against local or cloud-hosted project files, according to the README. Its listed areas include project files, offline references, conform and relink checks, grade transfer, Fusion, audio planning, Fairlight routing, project databases, pipeline stages, capability inspection, deliverable checks, and media operations.

## Setup and configuration

Install the managed copy and run the installer with:

```bash
npx davinci-resolve-mcp setup
```

The installer creates a Python virtual environment, detects Resolve paths, and can configure clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, Continue, Cline, Roo Code, OpenCode, Codex CLI, and JetBrains IDEs. A source checkout can instead be installed with `python install.py`.

The bridge installer is run separately with `python scripts/install_resolve_bridge.py`. Its default configuration file is `~/.config/davinci-resolve-mcp/bridge.json`. Set `DAVINCI_RESOLVE_BRIDGE_CONFIG` in both the installer and MCP client environment to use another path. Setting `DAVINCI_RESOLVE_BRIDGE=1` forces bridge transport.

The local control panel starts from the repository root with `venv/bin/python -m src.control_panel` and listens on loopback with a per-launch access token.

## Tools and capabilities

Capabilities described by the project include:

- Live Resolve editing and project control
- Media pool organization and media analysis
- Render setup and review markers
- Color grading, Fusion, and Fairlight workflows
- Offline Resolve file and project-database operations
- Local browser-based state inspection and analysis review

## Limitations and notes

The live integration requires DaVinci Resolve 18.5 or later and Python 3.10 or later, based on the project metadata. Studio users need external scripting enabled. Free-edition users depend on the in-app bridge, which relies on a Resolve behavior that Blackmagic could change. The bridge is loopback-only and uses HMAC-signed requests with one-use nonces.

On macOS, Resolve's Python discovery expects either `PYTHON3HOME` or `/usr/local/bin/python3`; Resolve launched from the Dock does not inherit ordinary shell exports. The README reports validation on macOS, Windows 11, and Linux, while noting that some Windows paths were initially unverified before user reports confirmed them.

The project checks GitHub releases for updates on a best-effort, throttled basis and does not block MCP startup for an update prompt.

_Full upstream README: https://allmcps.com/mcp/samuelgursky-davinci-resolve-mcp/readme_