🟡 Intermediate ⚙️ Type: MCP Server / Video Editing AI 💸 Free & Open Source (MIT) ⭐ 1,200+ GitHub Stars
What is DaVinci Resolve MCP?
The DaVinci Resolve MCP Server (created by Samuel Gursky) is a revolutionary bridge that connects your video editing software directly to modern AI assistants like Claude Desktop, Cursor, or the Gemini CLI. It acts as an expert translator, turning your natural language commands into direct actions inside DaVinci Resolve.
Instead of manually clicking through menus to organize bins, create multicam sequences, or render timelines, you simply tell your AI: “List all projects and open ‘My Film’,” or “Create a timeline called ‘Assembly Cut’ from all clips in the current bin.” The Model Context Protocol (MCP) standardizes this communication, granting the AI read-and-write access to your timeline, media pool, and project metadata.
This completely reimagines post-production workflows. You no longer need to be a Python scripting expert to automate tedious video editing tasks; you just need to be good at describing what you want the software to do.
Who is it for?
- Professional Video Editors and Colorists who want to automate repetitive project setups, bin organizations, and media imports using voice or text prompts.
- Post-Production Supervisors managing massive project files who need an AI assistant to quickly summarize timeline structures or locate specific assets.
- Developers looking for an out-of-the-box MCP integration to build custom video automation apps without writing the API wrappers from scratch.
- Content Creators seeking to speed up their rough cuts by commanding their AI to build assembly timelines instantly.
What makes it special?
- 100% API Coverage — It doesn’t just do basic tasks. It exposes over 324 features of the Resolve API, meaning you can control editing, color grading, audio mixing, and delivery entirely through natural language.
- Universal Setup Script — The developer included an incredible automated setup tool that detects your Python paths, creates the virtual environment, and automatically configures your AI clients (like Claude, Cursor, VS Code, and Windsurf).
- Local Control Panel — Includes a built-in web dashboard running on localhost where you can manually monitor the MCP server, run analysis jobs, and index your project metadata without opening a chat window.
- Compound vs. Granular Modes — Offers a “Compound” mode for most users (grouping related operations to save on AI token usage) and a “Granular” mode for power users who want one strict MCP tool per exact API method.
Requirements before you start
Because this tool interfaces deeply with professional software, you must meet a few strict prerequisites:
- DaVinci Resolve STUDIO — Critical: The free version of Resolve does not support the external Python scripting API. You must have the paid Studio license (version 18.0 or later).
- Python 3.10+ — Installed system-wide. On Windows, make sure you selected “Add to PATH” and “Install for all users” during setup so Resolve can find the registry keys.
- Node.js / npm (Optional but recommended) — Required if you want to use the easy one-line
npxsetup launcher. - An MCP Client — An AI application like Claude Desktop or Cursor installed to send the commands.
Step-by-step installation
Step 1 — Enable External Scripting in Resolve
Before doing anything with code, you must give Resolve permission to listen to outside scripts:
- Open DaVinci Resolve Studio.
- Go to Preferences > System > General.
- Find the setting External scripting using and change it from None to Local.
- Save and leave DaVinci Resolve running in the background.
Step 2 — Run the Automated Setup
Open your terminal or command prompt. The easiest way to install and configure everything is using the official NPM launcher, which handles all the heavy lifting:
npx davinci-resolve-mcp setup(If you prefer to install from source manually, you can clone the GitHub repo and run python install.py inside the folder).
Step 3 — Follow the Configuration Wizard
The setup script will automatically download the Python dependencies, create a virtual environment, and then ask which AI clients you have installed (e.g., Claude Desktop, Cursor). Select your preferred clients, and the script will automatically edit their mcp.json config files to point to the server.
Step 4 — Start Your First Edit
With Resolve still open in the background, open your configured AI assistant (like Claude Desktop). Ensure the “DaVinci Resolve” tool is active, and try sending your first command:
"List all the timelines in my current project and tell me their frame rates."If the setup was successful, the AI will reach into Resolve, read the data, and reply to you instantly!
Common errors and fixes
| Error | What it means | How to fix it |
|---|---|---|
Unable to connect to DaVinci Resolve | The server cannot find the software, or the API is locked. | Ensure DaVinci Resolve Studio is currently open and running. Verify that you enabled “Local” external scripting in the System Preferences (Step 1). |
Python not found (Windows) | DaVinci Resolve looks for specific Windows Registry keys to find Python, which are missing. | Reinstall Python from python.org. Check the box at the bottom that says “Add Python to PATH” and choose the “Install for all users” option to write the correct system keys. |
| Cursor/Claude says “Tool not found” | Your AI client’s MCP configuration file is pointing to the wrong folder path. | Rerun the npx davinci-resolve-mcp setup command to let it re-write the config file automatically, or manually check your claude_desktop_config.json to ensure the absolute paths are correct. |
Free vs Paid comparison
| Feature | DaVinci MCP (Free Open Source) | Proprietary NLE Automation Plugins |
|---|---|---|
| Cost of Plugin/Server | $0 (Free forever) | Usually $50–$200+ |
| Underlying Software Requirement | ⚠️ Requires paid DaVinci Resolve Studio ($295) | Varies (Premiere Pro subscription, etc.) |
| Language Model Flexibility | ✅ Use any MCP client (Claude, Gemini, local LLMs) | ❌ Locked to the developer’s chosen API |
| API Coverage | ✅ 100% full application access | Limited to specific niche features |
Bottom line: This open-source server is a masterpiece for post-production pros, but the barrier to entry is the mandatory $295 DaVinci Resolve Studio license. If you already own Studio, this tool gives you enterprise-grade AI automation for free. If you only use the free version of Resolve, you cannot use this tool and will need to rely on manual editing.
Alternatives — 3 similar tools
1. Positronikal / davinci-mcp-professional
An enterprise-focused hard fork of this exact project. It offers standalone Windows executables (so you don’t need to install Python at all) and strips out the Node.js dependencies in favor of pure uv package management. Great for studio IT admins deploying the server to dozens of editor workstations.
🔗 github.com/Positronikal/davinci-mcp-professional
2. Firecut AI (Premiere Pro)
If you use Adobe Premiere instead of Resolve, Firecut is a highly polished commercial plugin. It isn’t an open MCP server, but it provides dedicated AI buttons to automatically cut silences, add b-roll, and create chapters without writing any prompts.
3. AutoPod
Another commercial powerhouse for Adobe users, explicitly designed to automate multi-camera podcast editing. While DaVinci MCP can do this if prompted correctly, AutoPod is a specialized, one-click solution that requires zero LLM prompting, though it comes with a monthly fee.
🚀 Want more free AI tools like this?
We find, test, and write setup guides for the best free and open-source AI tools — so you don’t have to dig through GitHub yourself.Browse Free AI Tools at globalaiforce.com/shop →
📸 Follow us for daily AI tool tips and tutorials: instagram.com/globalaiforce