PostHog Ecosystem

PostHog is the product analytics and feature-flag route when the task is about user behavior, events, funnels, experiments, autocapture, session replay, or LLM analytics. Source: Ecosystem Capability Coverage, 2026-06-27; Skill Resolver, 2026-06-27

When To Use This

Use PostHog when product behavior needs event evidence, feature flag control, experiment analysis, session replay, or analytics instrumentation. Use Sentry for exceptions and Datadog for infra/service telemetry. Source: Skill Resolver, 2026-06-27

Decision Matrix

Core Model

PostHog is product telemetry. It should answer "what did users do and under which flag/experiment?" It is not the primary route for stack traces or infrastructure metrics. Source: Observability Architecture, 2026-06-27

Routing Summary

Preferred order: PostHog MCP for product data queries, plugin skills for instrumentation and analytics workflows, then app tests for event emission. The local coverage audit tracks skills.sh and MCP evidence but no first-party CLI seed. Source: Ecosystem Capability Coverage, 2026-06-27

Failure Modes

• Emitting events without naming schema or ownership.
• Mixing product analytics with exception tracking.
• Shipping flags without rollback and owner.
• Treating autocapture as enough for decision-grade analytics.

Verification Path

Confirm project/environment, run a read-only HogQL query where possible, and prove new instrumentation by capturing a sample event with the expected properties. Source: Skill Resolver, 2026-06-27

---

Timeline

• 2026-06-27 | Canonical ecosystem page created to give PostHog product analytics its own route instead of burying it under generic observability. Source: User request, 2026-06-27; Ecosystem Capability Coverage