CapCut MCP
Enables AI assistants like Claude Code to build and edit CapCut video projects locally by creating drafts, adding media and effects, and saving projects that open in CapCut desktop.
README
CapCut MCP — edit CapCut videos with an AI agent
This is a local MCP server that lets an AI assistant (like Claude Code) build and edit CapCut video projects for you — creating drafts, adding video/audio/images/text/subtitles/effects, and saving projects that open straight in the CapCut desktop app.
It runs entirely on your machine and makes no outbound calls to do its work. (See Security.)
The easy way: let the AI install it
- Install CapCut desktop and Claude Code, and make sure you have Python 3.10+.
- Clone this repo and open it in Claude Code:
git clone <your-repo-url> capcut-mcp cd capcut-mcp claude - Tell Claude: "Set this up." It will read
CLAUDE.md/AGENTS.mdand install, configure, and register everything itself. - Restart Claude Code. You now have
capcuttools — ask it to edit a video.
The manual way
python install.py
This creates a virtual environment, installs dependencies, finds your CapCut
drafts folder, writes a local config.json, verifies the server, and registers
it with Claude Code. Full manual steps are in AGENTS.md.
What it can do
Once registered, the assistant can call tools including:
create_draft— start a new CapCut projectadd_video,add_audio,add_image— place media on the timelineadd_text,add_subtitle— captions and titlesadd_effect,add_sticker,add_video_keyframe— effects and animationget_video_duration— inspect medialist_drafts,load_draft,save_draft— open and save projects
Drafts are written into CapCut's projects folder, so they show up in the CapCut app automatically.
Requirements
- Python 3.10+
- CapCut desktop (Windows or macOS)
- An MCP client — Claude Code recommended
Security
This is a hardened redistribution. The upstream project it derives from contacted two remote endpoints; both have been disabled so the tool stays fully local:
- a remote draft-retrieval cloud function, and
- a remote default
draft_domain.
config.json (which can hold cloud-storage keys if you opt in) and .venv are
git-ignored and never published. Remote draft upload is off by default.
See NOTICE for attribution and the full list of changes.
License
Licensed under the Apache License 2.0 (see LICENSE). This
project is a modified redistribution of
CapCutAPI; see NOTICE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.