Neura Market
samuelgursky/davinci-resolve-mcp logo

samuelgursky/davinci-resolve-mcp

Free

DaVinci Resolve 的 MCP 服务器集成，提供视频编辑、调色、媒体管理和项目控制的强大工具。

FreeFree tier
Inputs: textOutputs: text
Type
Open Source
Visit Website GitHub (40 stars)

About samuelgursky/davinci-resolve-mcp

A Model Context Protocol (MCP) server that lets AI assistants control DaVinci Resolve Studio through the official Scripting API. 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. Includes a local browser control panel for inspecting Resolve state, running source-safe analysis, drilling into clips and shots, and editing analysis output inline. Supports multiple AI coding assistants and IDEs (Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, Continue, Cline, Roo Code, OpenCode, JetBrains). Two server modes: Compound (34 tools, grouped operations for low context) and Full/Granular (341 tools, one per API method).

Key Features

Full DaVinci Resolve Scripting API coverage
Guarded workflow helpers for editing, media pool, render, review, grading, Fusion, Fairlight, project lifecycle
Local browser control panel for inspecting state and analyzing media
Source-safe media analysis (read-only, no modification)
Support for multiple AI assistants and IDEs: Claude, Cursor, VS Code, Windsurf, Zed, Continue, Cline, Roo Code, OpenCode, JetBrains
Two server modes: Compound (34 tools) and Full/Granular (341 tools)
Easy setup via npx or Python installer, with automatic client configuration
Cross-platform: works on Windows, macOS, Linux (where Resolve runs)

Pros & Cons

Pros
  • Completely open source and free
  • Comprehensive API coverage with guarded safe helpers
  • Easy setup via npx with automatic client configuration
  • Local control panel provides visibility into Resolve state
  • Supports multiple popular AI coding assistants and IDEs
  • Two server modes allow choosing between compact or detailed toolset
  • Active community (1.8k stars, 200 forks on GitHub)
Cons
  • Requires DaVinci Resolve Studio (paid software) – does not work with free version
  • Only works on the same machine where Resolve is installed (local MCP server)
  • Setup requires manual enabling of external scripting in Resolve preferences
  • Server may have a learning curve for users unfamiliar with MCP protocol
  • Granular mode uses many tools which may exceed token limits on some models

Best For

Automate video editing tasks via AI assistant promptsMedia pool organization and clip managementColor grading and Fusion composition controlRender setup and export automationReview marker placement and project lifecycle managementFairlight audio editing integrationExtension authoring and custom scriptingNon-destructive media analysis and shot inspection

FAQ

How do I install the DaVinci Resolve MCP server?
Run npx davinci-resolve-mcp setup in your terminal. The installer will create a virtual environment, detect Resolve paths, and configure supported IDEs automatically. Alternatively, clone the repository and run python install.py.
What prerequisites are needed?
DaVinci Resolve Studio must be installed and running. In Resolve, go to Preferences > General > External scripting and set it to 'Local'. The server works on Windows, macOS, and Linux.
What server modes are available and how do they differ?
Two modes: Compound mode (default, 34 tools) groups related operations behind action parameters to keep context usage low. Full/Granular mode (341 tools) provides one MCP tool per Resolve API method. Use --full flag to switch.
Can I view or analyze media without modifying it?
Yes, the control panel and analysis features are source-safe. They perform read-only operations and never alter your project or media files.
Which AI assistants and IDEs are supported?
Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, Continue, Cline, Roo Code, OpenCode, and JetBrains IDEs are supported. The installer can automatically configure these clients.
Is this project free to use?
Yes, it is completely free and open source under the MIT license.