DaVinci Resolve MCP Server provides a Python-based MCP server and an optional Node-based advanced server for editing, media analysis, and project control within DaVinci Resolve. It includes a local control panel, multiple server modes, and optional extras for extended capabilities.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
A Model Context Protocol (MCP) server that lets AI assistants control DaVinci Resolve Studio through the official Scripting API. It provides full API coverage plus guarded workflow helpers for editing, media pool organization, render setup, review markers, grading, Fusion, Fairlight, project lifecycle tasks, extension authoring, and source-safe media analysis. A local browser control panel ships with the server for inspecting Resolve state, running source-safe analysis, drilling into analyzed clips and shots, and editing analysis output inline. See the Quick Start and Installation sections for setup details.
How it works
The repository ships a Python MCP server (compound mode by default) and an optional advanced MCP server (Node) that can operate alongside. The Python server drives a live Resolve instance via the scripting API, while the advanced server handles file-level operations in scenarios where Resolve is not running. The package can be installed via npm-driven setup or a source install with Python. A local control panel runs via a separate Python module.
Getting started
Quick Start commands from the README:
npx davinci-resolve-mcp setup
For source installs:
git clone https://github.com/samuelgursky/davinci-resolve-mcp.git
cd davinci-resolve-mcp
python install.py
Local control panel launch:
venv/bin/python -m src.control_panel
Platform-paths and manual setup details are available in docs/install.md. The installer checks the latest GitHub release for MCP updates and can prompt or auto-update when using the standard installer flow.
Recent releases
Latest releases include:
- v2.76.0 (2026-08-04): Three AAF conform-fidelity fixes found by placing a full 83-minute Avid turnover and cross-checking it against Resolve's own native import of the same file. Fixed: AAF transitions.
- v2.75.0 (2026-08-04): Motion Control speed recovery. The offline AAF reader now recovers retime ratios instead of only flagging them.
- v2.73.2 (2026-08-04): Two honesty fixes in the conform path;
detect_missing_mediagains two additive response fields. - v2.73.1 (2026-08-04): Packaging fix; npm package shipped the AAF reader's Node half without its Python half.
- v2.73.0 (2026-08-04): AAF multi-layer timelines support in offline reader.
Notes: The README excerpt provided is truncated; the above summarizes each listed release entry verbatim.
Traction
Stars: 1987 forks: 229 open_issues: 2
Behind the repo
This section is not applicable as there is no linked startup/company information in the provided facts.
Caveats
License: MIT Language: Python Created: 2025-03-18 Last push: 2026-08-04 Additional notes:
- The project targets Python 3.10+ as per README badges.
- The repository includes an optional Node-based MCP server (davinci-resolve-advanced-mcp) and an npm-based installation flow.
- The README mentions that the core install requires Python, ffmpeg, and the Resolve scripting API; additional features may require extra dependencies (ffmpeg, numpy, librosa, openai-whisper, open_clip_torch, transformers, opencv-python) according to the extras table.