Frequently Asked Questions about Goatcounter MCP Server

Q: What is the Goatcounter MCP Server?

A: The Goatcounter MCP Server acts as a bridge, enabling AI models to interact with Goatcounter analytics through a standardized interface.

Q: What is MCP?

A: MCP is an open protocol that standardizes how applications provide context to LLMs.

Q: How does the Goatcounter MCP Server handle API rate limits?

A: It automatically retries requests with exponential backoff, prioritizing the X-Rate-Limit-Reset header if provided by the API.

Q: What environment variables are required to run the server?

A: GOATCOUNTER_CODE (your site code) and GOATCOUNTER_API_KEY (your API token).

Q: How do I install the Goatcounter MCP Server for use with Claude Desktop?

A: The recommended way is to use Smithery: npx -y @smithery/cli install @rafaljanicki/goatcounter-mcp-server --client claude. Alternatively, you can install from PyPI or source.

Q: Can I list the available tools even without configuring API credentials?

A: Yes, the server supports lazy initialization of the API client, allowing you to list tools before setting up credentials.

Q: What tools are available through the Goatcounter MCP Server?

A: Tools for getting user info, listing sites, listing paths, and getting stats on totals, hits, referrers, browsers, systems, sizes, and locations.

Q: How do I configure Claude Desktop to use the Goatcounter MCP Server?

A: You need to edit the claude_desktop_config.json file, adding the server configuration with the correct command, arguments, and environment variables.

Q: Where is the claude_desktop_config.json file located?

A: On Windows: %APPDATA%Claudeclaude_desktop_config.json. On macOS: ~/Library/Application Support/Claude/claude_desktop_config.json.

Q: Can I use the Goatcounter MCP Server with UBOS?

A: Yes! Integrating the Goatcounter MCP Server with UBOS allows you to seamlessly incorporate web analytics data into your AI Agent workflows.

Q: What are the benefits of using Goatcounter MCP Server with UBOS?

A: Centralized AI Agent management, enhanced data connectivity, simplified AI Agent development, scalable AI Agent infrastructure and improved collaboration.

Q: Where can I generate an API key for Goatcounter?

A: In your Goatcounter site, go to Settings -> API tokens.

Q: What is FastMCP?

A: FastMCP is a library used to build the MCP server, enabling communication between AI models and external data sources.