{ "markdown": "## CutPro MCP\n\n[![npm version](https://img.shields.io/npm/v/@cutpro/mcp?style=flat-square&color=7C3AED)](https://www.npmjs.com/package/@cutpro/mcp)\n[![MCP Registry](https://img.shields.io/badge/MCP_Registry-io.github.getcutpro%2Fcutpro-7C3AED?style=flat-square)](https://registry.modelcontextprotocol.io)\n[![smithery badge](https://smithery.ai/badge/contact-8lma/cutpro)](https://smithery.ai/servers/contact-8lma/cutpro)\n\nA Model Context Protocol (MCP) server that turns long videos into viral clips with AI. It exposes the full [CutPro API](https://cut.pro/docs/api-reference) 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.\n\n### Key features\n\n- **End to end**. All 34 v1 endpoints as tools: workspace, balance, videos, clipping, clips, templates, renders, posts and connections.\n- **Token efficient**. Results are compact and projected to the fields that matter; `list_clips` is rating sorted, capped, and omits long signed URLs unless asked.\n- **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.\n\n## Getting started\n\n### Requirements\n\n- Node.js 18 or newer.\n- A CutPro account on the **Pro** plan and an API key. Generate one at [cut.pro/studio/me/api-keys](https://cut.pro/studio/me/api-keys).\n- An MCP-compatible client.\n\n### Standard config\n\nMost clients use the same JSON. Add your API key under `env`:\n\n```json\n{\n \"mcpServers\": {\n \"cutpro\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@cutpro/mcp\"],\n \"env\": { \"CUTPRO_API_KEY\": \"\" }\n }\n }\n}\n```\n\n### Install\n\n[\"Install](https://insiders.vscode.dev/redirect?url=vscode%3Amcp%2Finstall%3F%257B%2522name%2522%253A%2522cutpro%2522%252C%2522command%2522%253A%2522npx%2522%252C%2522args%2522%253A%255B%2522-y%2522%252C%2522%2540cutpro%252Fmcp%2522%255D%257D)\n[\"Install](https://cursor.com/en/install-mcp?name=cutpro&config=eyJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkBjdXRwcm8vbWNwIl19)\n[\"Install](https://lmstudio.ai/install-mcp?name=cutpro&config=eyJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkBjdXRwcm8vbWNwIl19)\n[\"Install](goose://extension?cmd=npx&arg=-y&arg=%40cutpro%2Fmcp&id=cutpro&name=CutPro&description=AI%20video%20clipping%20via%20the%20CutPro%20API)\n\nAfter installing via a button, add your `CUTPRO_API_KEY` to the server's `env`.\n\n
\nClaude Code\n\n```bash\nclaude mcp add cutpro --env CUTPRO_API_KEY= -- npx -y @cutpro/mcp\n```\n
\n\n
\nClaude Desktop\n\nAdd to `claude_desktop_config.json` (Settings, Developer, Edit Config):\n\n```json\n{\n \"mcpServers\": {\n \"cutpro\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@cutpro/mcp\"],\n \"env\": { \"CUTPRO_API_KEY\": \"\" }\n }\n }\n}\n```\n
\n\n
\nCursor / Windsurf / VS Code (manual)\n\nAdd the standard config above to the client's MCP settings (`mcp.json` / `mcpServers`).\n
\n\n
\nCline\n\nOpen the MCP Servers panel, choose Configure, and add the standard config above.\n
\n\n
\nGemini CLI\n\n```bash\ngemini mcp add cutpro npx -y @cutpro/mcp -e CUTPRO_API_KEY=\n```\n
\n\n
\nCodex\n\nAdd to `~/.codex/config.toml`:\n\n```toml\n[mcp_servers.cutpro]\ncommand = \"npx\"\nargs = [\"-y\", \"@cutpro/mcp\"]\nenv = { \"CUTPRO_API_KEY\" = \"\" }\n```\n
\n\n
\nChatGPT and Claude.ai (hosted, no install)\n\nUse the hosted server. Add a custom connector pointing to:\n\n```\nhttps://mcp.cut.pro\n```\n\nYou authorize with your CutPro API key on a consent page (OAuth), so no local setup is needed.\n
\n\n## Configuration\n\nThe server is configured with environment variables.\n\n| Variable | Description | Required |\n| --- | --- | --- |\n| `CUTPRO_API_KEY` | Your CutPro API key (Pro plan). | Yes (stdio) |\n| `CUTPRO_WORKSPACE_ID` | Selects the workspace for multi-workspace keys. | No |\n| `CUTPRO_API_URL` | Override the API base URL. Defaults to `https://api.cut.pro/api/v1`. | No |\n\n
\nSelf-hosting the remote (Streamable HTTP + OAuth)\n\n| Variable | Description |\n| --- | --- |\n| `MCP_TRANSPORT=http` / `PORT` | Serve Streamable HTTP at the root instead of stdio. |\n| `MCP_OAUTH=1` | Enable the full OAuth 2.1 layer (discovery, DCR, PKCE) for browser clients. |\n| `MCP_PUBLIC_URL` | Public endpoint, e.g. `https://mcp.cut.pro`. Its origin becomes the OAuth issuer. |\n| `MCP_REDIS_URL` | Back OAuth state with Redis so it survives restarts and scales across instances. |\n\n```bash\nMCP_TRANSPORT=http PORT=8787 MCP_OAUTH=1 \\\nMCP_PUBLIC_URL=https://mcp.cut.pro MCP_REDIS_URL=redis://127.0.0.1:6379 \\\nnpx -y @cutpro/mcp\n```\n\nIn OAuth mode the user authorizes with their own API key on a consent page; the access token maps server side to that key. Without `MCP_REDIS_URL`, an in-memory store is used (single instance, state lost on restart).\n
\n\n## Tools\n\n
\nWorkspace and balance\n\n- **get_workspace**: the workspace this key resolved to, with plan and role.\n- **get_balance**: current credit balance.\n- **get_balance_history**: ledger of credits added and consumed.\n
\n\n
\nVideos and uploads\n\n- **list_videos**: your source video library.\n- **start_upload**: get a presigned URL to upload your own file (max 2 GB; .mp4/.mov/.webm/.mkv).\n- **complete_upload**: register a finished upload and get its credit cost.\n- **delete_video**: delete a source video and its submissions.\n
\n\n
\nClipping\n\n- **analyze_video**: preview metadata and credit cost of a public URL (free).\n- **submit_clipping**: start AI clipping (charges credits).\n- **list_submissions**: clipping jobs for a video.\n- **get_submission**: poll a submission until completed or failed.\n- **delete_submission**: delete a submission and its clips.\n
\n\n
\nClips and templates\n\n- **list_clips**: clips of a completed submission, rating sorted (URLs opt-in).\n- **apply_template**: apply an editing template to clips in bulk.\n- **delete_clip**: delete a single clip.\n- **list_templates**: editing templates to apply to clips.\n
\n\n
\nRenders\n\n- **render_clip**: render a clip to a final MP4.\n- **get_render**: poll a render until completed.\n- **get_render_download**: signed download URL of a completed render.\n- **cancel_render**: cancel or delete a render.\n- **get_render_limits**: render quota for the workspace.\n- **start_bulk_download** / **get_bulk_download**: bundle several renders into one download.\n
\n\n
\nPosts and connections\n\n- **create_post**: publish rendered clips to connected accounts (immediate or scheduled).\n- **list_posts** / **get_post** / **update_post** / **delete_post**: manage posts.\n- **publish_post**: trigger publishing now.\n- **retry_post_item** / **delete_post_item**: handle individual targets.\n- **list_connections** / **get_connection**: connected social accounts.\n
\n\nEach tool carries read-only / write / destructive annotations so clients can plan calls.\n\n## Links\n\n- Docs: [cut.pro/docs/api-reference/mcp](https://cut.pro/docs/api-reference/mcp)\n- npm: [@cutpro/mcp](https://www.npmjs.com/package/@cutpro/mcp)\n- MCP Registry: `io.github.getcutpro/cutpro`\n\n## License\n\nMIT\n", "bytes": 8001, "sha": "c6d48244656d4f5487a95b25bbaa5b815ca97b1292e1877971dff595938ce0f9", "repo_slug": "getcutpro/mcp", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/mcp_io_github_getcutpro_cutpro_6bbbc849/readme" }