posthog

Evaluate PostHog feature flags and record experiment events for pilot routing.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/edupazogle/agentic_cockpit --skill posthog-edupazogle
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: posthog
Source: https://github.com/edupazogle/agentic_cockpit/tree/main/.agents/skills/posthog
Command: npx skills add https://github.com/edupazogle/agentic_cockpit --skill posthog-edupazogle

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you gate features and run flag-based A/B experiments in the Agentic Cockpit while ensuring correct analytics attribution and EU-compliant tracking.

Core Features & Use Cases

  • Feature flags for safe rollouts: Gate pilot flow behavior behind a tenant- and pilot-scoped pilot:<id> flag for controlled deployments and kill-switches.
  • Experiments for comparing variants: Run experiments between two Langflow flow variants by mapping variant_id to PostHog flag variant keys and measuring decision outcomes.
  • Cohorts and event recording: Define cohort logic in PostHog while mirroring cohort predicates into Supabase for offline replay, and record a minimal event set like pilot.run.started and hitl.decision.made.
  • EU data residency correctness: Enforce using https://eu.i.posthog.com rather than default endpoints to satisfy residency requirements.

Quick Start

Integrate the posthog skill by configuring POSTHOG_HOST, POSTHOG_API_KEY, and POSTHOG_PROJECT_API_KEY, then evaluate flags server-side using distinct_id set to chatwoot_conversation_id with send_feature_flag_events enabled.

Frequently Asked Questions about posthog

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run A/B tests between different Langflow flow variants?

Run A/B tests by mapping a Langflow flow variant ID to PostHog flag variant keys and measuring the resulting decision outcomes. You evaluate flags server-side while recording minimal decision events to compare the performance of the two variants.

How does feature flag gating work for controlled pilot rollouts?

Feature flag gating works by evaluating tenant- and pilot-scoped flags, such as a pilot ID flag, to act as a kill-switch. This ensures you can safely control deployments and route behavior based on the specific pilot context evaluated server-side.

Do I need a specific PostHog host configuration for EU data residency?

Yes, EU data residency requires using the specific PostHog host endpoint rather than default endpoints. You must configure your setup to use the EU host to satisfy regional data residency and compliance requirements correctly.

What is the correct distinct ID for user stickiness in product analytics?

The correct distinct ID for user stickiness is the stable chatwoot conversation ID. Setting this ensures consistent user attribution across flag evaluations and experiment exposures throughout the entire pilot lifecycle.

Why are my feature flag events not appearing in experiment funnels?

Feature flag events may not appear in experiment funnels if send_feature_flag_events is not enabled. You must set this parameter to true during flag evaluation to guarantee the integrity of your experimentation analytics.

Can I use Supabase for offline replay of PostHog cohort logic?

Yes, you can mirror cohort predicates defined in PostHog into Supabase for offline replay. This allows you to maintain cohort-based routing logic locally while keeping your product analytics cohorts synchronized.