mcp -processon

This MCP server provides a bridge between structured Markdown content and ProcessOn mind maps. It exposes two tools via the MCP protocol: check() to verify that your ProcessOn API key is configured and to retrieve the current API endpoint, and createProcessOnMind(title, content) to transform Markdown into a ProcessOn mind map. To use it, start the server (for example via npx as shown in installation). Once running, you can call check() to confirm connectivity and then pass a Markdown string as the content with a title to createProcessOnMind to generate and publish a mind map on ProcessOn.

Prerequisites:\n- Node.js and npm (recommended Node.js 18+).\n- Optional: Python 3.8+ if you prefer the Python implementation, though the MCP server can run via npx without local Python setup.\n\nInstallation steps (recommended):\n1) Ensure you are authenticated with npm if you’re using a private registry.\n2) Run the MCP server via npx (no global install required):\n\nbash\nnpx @processon/mcp-server-processon\n\n3) Set required environment variables for ProcessOn access. Example (Unix):\nbash\nexport PROCESSON_API_KEY=YOUR_PROCESSON_API_KEY\n\nWindows (CMD):\ncmd\nset PROCESSON_API_KEY=YOUR_PROCESSON_API_KEY\n\n4) Optionally run with a custom base URL if you’re using a self-hosted ProcessOn API:\nbash\nexport BASE_URL=https://api.processon.com\n\n\nIf you prefer a global install:\n1) Install globally:\nbash\nnpm install -g @processon/mcp-server-processon\n\n2) Run:\nbash\n@processon/mcp-server-processon\n

Tips and notes:\n- The server reads ProcessOn configuration from environment variables. Required: PROCESSON_API_KEY. Optional: BASE_URL to override the official API endpoint.\n- The MCP interface exposes two tools: check() and createProcessOnMind(title, content). Use createProcessOnMind with a Markdown string where the root is the first level-1 heading and subheadings map to nested nodes (levels 2-6).\n- When using npx, the first run may install dependencies; ensure network access. The README mentions automatic Python dependency installation on first run if you choose the Python path.\n- For Cherry Studio or similar integrations, you can import the JSON snippet from the README to register the MCP server.\n- If you encounter authentication errors, double-check that PROCESSON_API_KEY is valid for the ProcessOn account attached to the API key.\n- You can also pass BASE_URL to point to a custom ProcessOn API endpoint if needed.