DaVinci Resolve MCP Server

What is DaVinci Resolve MCP Server?

An 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. The server ships with a local browser control panel for inspecting Resolve state and running analysis.

How to use DaVinci Resolve MCP Server?

Install via npx davinci-resolve-mcp setup or clone the repo and run python install.py. Before connecting, open DaVinci Resolve Studio and set Preferences > General > External scripting using to Local. The installer creates a virtual environment, detects Resolve paths, and can configure Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, Continue, Cline, Roo Code, OpenCode, and JetBrains IDEs. The compound server (default) provides 34 tools; the full granular mode provides 341 tools. An optional Node.js‑based advanced server (davinci-resolve-advanced-mcp) adds 18 offline tools for file‑level operations.

Key features of DaVinci Resolve MCP Server

• Full coverage of DaVinci Resolve Scripting API (100% API coverage claimed)
• Two server modes: compound (34 tools) and granular (341 tools)
• Local browser control panel for inspecting Resolve state and analysis results
• Source-safe media analysis with sidecar reports, transcription, and slate/clap detection
• Media pool, timeline editing, review annotations, color grading, and render helpers
• Extension authoring helpers for Fuse, DCTL, Lua, and Python scripts
• Optional advanced server for offline file-level operations (DRP, DRT, DRX, etc.)
• Installer automatically configures popular MCP clients and IDE tools

Use cases of DaVinci Resolve MCP Server

• Automate project creation, timeline building, and bin organization from AI prompts
• Run source-safe media analysis (transcription, frame analysis, sync event detection) and write results into Resolve metadata
• Validate and queue renders, set up review markers, and export review reports
• Perform grade surgery, CDL validation, and LUT export via natural language commands
• Use the advanced server to apply grades, conform, and run pipeline tasks without Resolve running (cloud or local)

FAQ from DaVinci Resolve MCP Server

What is the difference between the compound and full server modes?

The compound server (default, 34 tools) groups related Resolve operations behind action parameters to keep context usage low. The full granular server (341 tools) exposes one MCP tool per Resolve API method for power users who need fine-grained control.

What are the dependencies and runtime requirements?

The live server requires Python 3.10+, DaVinci Resolve Studio 18.5+, and the Resolve Scripting API (set to Local mode). The advanced server requires Node.js and can optionally use ffmpeg, sharp, and better-sqlite3 (checked via the capabilities tool). Both servers check for updates but never block startup.

How does the advanced server differ from the live server?

The live Python server drives a running Resolve via the scripting API. The advanced Node.js server reads and edits Resolve files (DRP/DRT/DRX) without needing Resolve running, enabling cloud or local offline operations like grade transfer, conform QC, and pipeline automation.

Can the server work offline?

The live server requires DaVinci Resolve to be running. The advanced server can work entirely offline on Resolve files. Analysis tools read source media but write reports only to sidecar, scratch, or project databases; camera originals are never modified.

How does the server handle source media safety?

The server treats camera originals and source media as immutable. Analysis tools read source files and write reports only to sidecar, scratch, or project databases. No source media is altered during any operation.