HomeProjectsArticlesPK BoardAbout
Explore Projects

PostHog

Active
GitHub Python NOASSERTION

Description

PostHog is a full-stack observability platform for AI agents and self-driving products, combining analytics, session replay, feature flags and error tracking.

Key Features

  • AI observability — Trace, token usage and diagnostic dashboards purpose-built for AI agents
  • Session replay — Capture complete agent execution event streams for replay and debugging
  • Feature flags & experiments — A/B testing and progressive rollouts coordinated with agent behavior
  • Multi-source capture — Unified error tracking, logs and analytics forming full context
  • MCP integration — Coordinate agent tasks from Slack, web, desktop or MCP
  • Self-hostable — Open-source core deployable locally or on private cloud

Use Cases

💡 Tracing every tool call and decision path of production AI agents
💡 Diagnosing root causes of abnormal agent behavior via session replay
💡 Running A/B experiments on agent behavior to compare models or prompts
💡 Triggering agent tasks from Slack and receiving execution results
💡 Building self-hosted observability platforms for compliance and data security

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (37.7k stars)

⚠️ Limitations

  • • High issue backlog (5.3k open issues)
  • • No clear open-source license

Tags

analytics session-replay feature-flags ai-observability experiments agent-diagnostics

Categories

📊 Observability ⚡ Agent Tools

Quick Start

# Start PostHog with Docker Compose
git clone https://github.com/PostHog/posthog.git
cd posthog
docker compose up -d

# Visit http://localhost:8000 to log in
Visit GitHub Visit Website View Docs

Related Projects

BAML

9.0k · Rust
Active A

BAML is an AI framework that adds engineering rigor to prompt engineering, offering type-safe prompt definitions, automatic testing, version management, and multi-model support across Python, TypeScript, Ruby, Java, C#, Rust, and Go.

prompt-engineeringtype-safellm-testing +2
  • · Type-safe prompt definition language that transforms prompt engineering into schema engineering
  • · Built-in IDE plugins (VS Code / JetBrains) for rapid prompt testing and iteration
  • · SAP (Schema-Aligned Parsing) algorithm for reliable structured output even without native tool-calling support
View full details → View on GitHub