Query MCP Analytics over PostHog MCP

PostHog's MCP server provides tools to query MCP Analytics. Use Claude Code, Cursor, or another MCP client to investigate tool usage and failures without opening the web app.

Before you start

1. Install MCP Analytics – install the SDK in your MCP server so tool calls are captured.
2. Connect the MCP server – follow the MCP server setup to connect PostHog to your client. MCP Analytics reads need mcp_analytics:read. Tool-quality queries also need query:read. Generating session summaries, reporting missing capabilities, and recomputing clusters need mcp_analytics:write.

What you can do here

Explore sessions

• Filter the list of MCP sessions
• Open a single session and read its tool calls in order
• Generate an intent summary for a session

Understand intent

• Retrieve intent clusters and their tool distribution
• Recompute clusters after you capture new data

Dig into tool quality

• Per-tool stats: volume, error rate, and latency
• Daily trend for a tool over time
• Recent failures, with error type and message
• Top users of a given tool
• Neighboring tools that get called alongside it
• Sample intents behind its calls
• Harness and client breakdown

Report missing capabilities

• Report a missing capability your agent needed but couldn't find
• Submit other tool feedback through the API

Example prompts

Point your agent at these once the tools are connected:

• "Which of my MCP tools has the highest error rate this week?"
• "Show me the last 10 failed calls to search_docs and what the errors were."
• "What are agents actually trying to do when they call create_issue? Summarize the intent clusters."
• "Which tools get called right before run_query?"

Related

• Explore the same data visually in the web app.
• See the full list of PostHog MCP tools in the MCP docs.
• Act on Self-driving reports about failing tools in PostHog Desktop.