This MCP server provides enterprise-grade analytics and productivity insights by integrating Rize.io's time tracking and productivity data with Claude Desktop and other AI assistants using the Model Context Protocol. It exposes a suite of tools for computing productivity metrics, custom analytics reports, and focus-session analysis. Core capabilities include get_productivity_metrics for comprehensive analytics, get_analytics_report for executive-style insights, get_productivity_summary for daily performance overviews, get_focus_sessions for in-depth session analysis, and project management tools like create_project and list_projects. The server also offers system utilities such as get_current_user and health_check to monitor status and usage. When used with Claude Desktop, you can configure the MCP server in Claude’s settings to point to this Node-based server (dist/index.js) and provide the necessary environment variables, allowing Claude to call these tools and receive structured responses with context for actionable AI-assisted work planning.

Prerequisites:

• Node.js (v14+ recommended) and npm/yarn installed
• Access to an API key for Rize.io (your_rize_api_key_here)

Installation steps:

1. Clone the repository git clone https://github.com/mariomosca/rizeio-mcp-server.git cd rizeio_mcp_server

2. Install dependencies npm install

3. Configure environment cp .env.example .env

   Edit .env to include your Rize.io API key and adjust settings as needed

   RIZE_API_KEY=your_rize_io_api_key

4. Build the project npm run build

5. Run the server (production) npm run start

6. Run in development with hot reload (optional) npm run dev

7. Run tests (optional) npm test

Environment variables and configuration:

• RIZE_API_KEY: Your Rize.io API key required for data access
• CACHE_MAX_SIZE and CACHE_TTL: Configure the LRU cache behavior for performance
• LOG_LEVEL: Set to info|debug|warn|error for verbosity
• RATE_LIMITING, RATE_LIMIT_MAX, RATE_LIMIT_WINDOW: Optional throttling controls Claude Desktop integration tips:
• In Claude’s config, point the MCP server to the built dist/index.js with command set to node and provide the required env vars (RIZE_API_KEY, etc.).
• Use the provided example JSON layouts to ensure proper formatting of mcpServers entries for Claude to invoke tools like get_productivity_metrics or get_focus_sessions. Common issues:
• Ensure the RIZE_API_KEY is valid; authentication errors will surface as API or validation errors.
• Build step must be completed before running in production; ensure dist/index.js exists.
• If using Docker or alternative runtimes, adapt the mcp_config accordingly.