PandaProbe is an open-source platform designed to help teams collaboratively trace, evaluate, monitor, and debug AI agents. It addresses the challenges of understanding and improving complex agent behaviors by providing essential tools for development and operational insight. The platform benefits AI engineers and researchers by offering a unified environment for agent lifecycle management, with integrations for popular frameworks like LangGraph, CrewAI, and the Claude Agent SDK.

How It Works

The architecture employs a microservices approach, featuring a FastAPI backend, Celery workers for asynchronous tasks, PostgreSQL for data persistence, and Redis for message brokering and caching. It distinguishes between a management plane (SDK/HTTP Client using Bearer tokens for authentication) and a data plane (API key for project context). Ingestion and evaluation jobs are enqueued to Redis and processed by Celery workers, enabling scalable and responsive operations. Evaluations leverage an LLM-as-a-judge mechanism for automated assessment.

Quick Start & Requirements

• Prerequisites: Docker must be installed and running.
• Primary install/run command:

  git clone https://github.com/chirpz-ai/pandaprobe.git
  cd pandaprobe
  ./start.sh
  

• Access: Dashboard available at http://localhost:3000, API reference at http://localhost:8000/scalar.
• Documentation: Links to client library documentation for quickstart and advanced integrations are mentioned but not directly provided in the README.

Highlighted Details

• Supports collaborative tracing, evaluation, monitoring, and debugging of AI agents.
• Integrates with popular agent frameworks like LangGraph, CrewAI, and the Claude Agent SDK.
• Offers both a managed cloud deployment with a free tier and a self-hostable Docker option.
• Provides a clear separation between management and data planes for robust API interactions.

Maintenance & Community

• Developed by the Chirpz AI team.
• Contact email for inquiries: sina@pandaprobe.com.
• No specific community channels (e.g., Discord, Slack) or roadmap links are detailed in the README.

Licensing & Compatibility

• License: Apache 2.0.
• Compatibility: The Apache 2.0 license is permissive, generally allowing for commercial use and integration within closed-source projects without significant restrictions.

Limitations & Caveats

Self-hosting requires Docker. No other specific limitations, alpha status, or known bugs are detailed in the provided README content.