A Model Context Protocol (MCP) server that turns long videos into viral clips with AI. It exposes the full CutPro API as tools, so an LLM can run the whole flow: analyze a video, clip the best moments, render the final MP4, and publish to TikTok, Instagram and YouTube.
Always in sync. Every v1 endpoint is a tool, generated from the live API spec: workspace, balance, videos and uploads, clipping, clips, channel and live monitoring, templates, renders, transcriptions, posts and connections.
Runs everywhere. stdio for local clients (Claude Code, Cursor, Claude Desktop, Windsurf, VS Code, Cline, Zed) and a hosted Streamable HTTP endpoint with OAuth for ChatGPT and Claude.ai.
Related MCP server: minimax-mcp-api
Node.js 18 or newer.
A CutPro account on the Pro plan and an API key. Generate one at cut.pro/studio/me/api-keys.
An MCP-compatible client.
Most clients use the same JSON. Add your API key under env:
{
"mcpServers": {
"cutpro": {
"command": "npx",
"args": ["-y", "@cutpro/mcp@latest"],
"env": { "CUTPRO_API_KEY": "<your-api-key>" }
}
}
}
After installing via a button, add your CUTPRO_API_KEY to the server's env.
claude mcp add cutpro --env CUTPRO_API_KEY=<your-api-key> -- npx -y @cutpro/mcp@latestAdd to claude_desktop_config.json (Settings, Developer, Edit Config):
{
"mcpServers": {
"cutpro": {
"command": "npx",
"args": ["-y", "@cutpro/mcp@latest"],
"env": { "CUTPRO_API_KEY": "<your-api-key>" }
}
}
}Add the standard config above to the client's MCP settings (mcp.json / mcpServers).
Open the MCP Servers panel, choose Configure, and add the standard config above.
gemini mcp add cutpro npx -y @cutpro/mcp@latest -e CUTPRO_API_KEY=<your-api-key>Add to ~/.codex/config.toml:
[mcp_servers.cutpro]
command = "npx"
args = ["-y", "@cutpro/mcp@latest"]
env = { "CUTPRO_API_KEY" = "<your-api-key>" }Use the hosted server. Add a custom connector pointing to:
https://mcp.cut.proYou sign in to cut.pro, pick the workspace and approve the connection, so no API key and no local setup are needed. Any CutPro plan can connect; jobs use the workspace's credits. Disconnect at any time in Settings, under Connected apps.
The plugin bundles the hosted server with a skill that teaches Claude the CutPro workflows:
claude plugin marketplace add getcutpro/mcp
claude plugin install cutpro@cutproThe server is configured with environment variables.
Variable | Description | Required |
| Your CutPro API key. | Yes (stdio) |
| Selects the workspace for multi-workspace keys. | No |
| Override the API base URL. Defaults to | No |
Variable | Description |
| Serve Streamable HTTP at the root instead of stdio. |
| Enable the full OAuth 2.1 layer (discovery, DCR, PKCE) for browser clients. |
| Public endpoint, e.g. |
| Back OAuth state with Redis so it survives restarts and scales across instances. |
| Where the consent screen lives. Defaults to |
| Domain verification token from the OpenAI plugin portal, served at |
MCP_TRANSPORT=http PORT=8787 MCP_OAUTH=1 \
MCP_PUBLIC_URL=https://mcp.cut.pro MCP_REDIS_URL=redis://127.0.0.1:6379 \
npx -y @cutpro/mcp@latestIn OAuth mode the user signs in on cut.pro and approves the app there; cut.pro creates a key for that app and workspace, and the access token maps server side to it. Clients can register dynamically (DCR) or present a Client ID Metadata Document (CIMD). Without MCP_REDIS_URL, an in-memory store is used (single instance, state lost on restart).
The tools are built from the API's own OpenAPI spec (/api/v1/openapi.json), so they always match the API reference: one tool per endpoint, named after its operationId in snake case (analyzeVideo is analyze_video, createSubmission is create_submission), with the endpoint's own parameters, descriptions and responses. A new endpoint becomes a tool within ten minutes of reaching the API, with no new release of this package.
The usual flow: analyze_video, create_submission, get_submission until it completes, list_clips, render_clip, get_render and download_render.
Annotations come from the spec too: GET is read-only, PUT, PATCH and DELETE are destructive, and a route declares more in its x-mcp extension (spending credits, publishing to a social network, reaching an arbitrary public URL). Routes that need a file uploaded from disk are listed only by the local (stdio) server, since a chat client cannot send bytes.
npm: @cutpro/mcp
MCP Registry: io.github.getcutpro/cutpro
MIT
MCP server for Clipkit — gives AI agents a video toolbox via the Clipkit schema.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
MCP server for QPost — lets AI agents publish video and image posts to YouTube, TikTok, Instagram.
Build, run, schedule, and publish AI video pipelines to YouTube and TikTok from any MCP client.
