Connects AI coding clients to CartCut’s live timeline for transcript cuts, captions, trims, motion, transitions, and effects.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent — or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag — we're steadily working through the catalog.
💡 Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by CartCut.
CartCut MCP server exposes an open CartCut project’s live timeline to an MCP-compatible AI client. The connection is intended for editing video rather than general file management or rendering automation. An agent can work from natural-language instructions to cut material from a transcript, add subtitles, trim or restage clips, and apply motion, transitions, and effects.
Edits are applied to the project inside CartCut and enter the application’s normal undo history. This means changes can be reversed with CartCut’s undo command, including the macOS shortcut mentioned in the project documentation. The underlying editor supports layer-based composition, text, audio mixing, keyframes, effects, and common media formats, but the README specifically describes the MCP connection in terms of timeline editing.
CartCut MCP server acts as the bridge between an MCP client and the CartCut desktop application. CartCut provides a connection command from its in-app lightning icon. That command contains a bearer token; the UUID following Bearer is the token used by the bridge.
For Claude Code, the repository provides a plugin containing the editing skill and connection setup. A separate skills CLI command can install only the skill, after which the MCP connection must be added using the command shown by CartCut. The documentation advises choosing one of these Claude Code routes rather than installing both.
Claude Desktop, Cursor, and Codex use the @cartesiancs/cartcut-mcp package. Running it with npx -y starts the bridge, which locates the token automatically after the client has been configured with the connection command.
Install the CartCut desktop application and open the project you want the agent to edit. Click the lightning icon at the bottom right of the CartCut interface and copy the connection command. For the package-based route, configure the MCP client to run:
The repository also documents these Claude Code options:
or:
The second command installs the skill only, so it still requires the MCP connection command from CartCut. The README points to the plugin documentation for connection troubleshooting.
CartCut MCP server supports agent-directed operations including:
The broader CartCut editor also lists AI auto-captioning with Whisper, keyframe animation, audio and screen recording, chroma key, blur, drawing, and FFmpeg-based rendering. The provided MCP documentation does not state that every listed editor feature is exposed as a separate MCP tool, so those capabilities should not be assumed to be directly callable through the bridge.
No reviews yet — be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/cartcut)<a href="https://allmcps.com/mcp/cartcut"><img src="https://allmcps.com/api/badge/cartcut?style=directory" alt="CartCut on AllMCPs" /></a>