posthog Plugin
Plugin Claude Code Monitoring Monitoring & ObservabilityData Engineering & AnalyticsAI Agents & AI App DevelopmentThe explanation below is AI-generated. Please verify it against the sources.
The posthog plugin connects Claude Code to PostHog's product analytics platform via the Model Context Protocol (MCP), letting you query analytics, manage feature flags, and debug errors directly from your editor. According to the README, it provides access to 27+ PostHog tools across categories such as feature flags, experiments, insights, dashboards, error tracking, and LLM analytics, plus 30+ bundled skills that activate automatically for matching requests. Installation is via claude plugin install posthog, followed by OAuth authentication through the /mcp command. The posthog homepage notes that the underlying MCP server is a free, hosted endpoint (at mcp.posthog.com), though tools that use LLMs internally may be billed as PostHog AI spend and are only available when AI data processing is enabled in organization settings. Self-hosted PostHog instances are supported by setting the POSTHOG_MCP_URL environment variable.
Overview
PostHog is a product analytics platform covering analytics, feature flags, experiments, and error tracking; this plugin exposes its tools and bundled skills to Claude Code via MCP.
What you can do with posthog
- Query analytics and insights (e.g., "How many users signed up this week?")
- Create and manage feature flags (e.g., roll out to 50% of users)
- View and analyze experiments/A-B tests, including results
- Create dashboards and add insights to them
- View, triage, and debug errors, including stack traces
- Track LLM/AI costs and usage via LLM analytics
- Search PostHog documentation from your editor
- Optionally send Claude Code session data to PostHog LLM Analytics
Sources
Original description (English)
Access PostHog analytics, feature flags, experiments, error tracking, and insights directly from Claude Code.
History of posthog
- Plugin Added posthog