llm-tracing

Python SDK for Agent AI Observability, Monitoring and Evaluation Framework. Includes features like agent, llm and tools tracing, debugging multi-agentic system, self-hosted dashboard and advanced analytics with timeline and execution graph view

Open-source, end-to-end platform for evaluating, observing, and improving LLM and AI agent applications. Tracing · Evals · Simulations · Datasets · Gateway · Guardrails. Self-hostable. Apache 2.0.

AgentInsight Python SDK provides a Python client for the AgentInsight platform, supporting LLM application observability, tracing, evaluation, and prompt management.

An exporter of openclaw traces to langfuse using logs

A Streamlit-based AI assistant generates custom Streamlit app code from user-provided images or text using the Google Gemini model.

🤖 Build advanced AI agents with a collection of production-ready applications using modern frameworks for single and multi-agent systems.

Single-binary, MLflow-API-compatible experiment tracker with first-class LLM trace support. 143× faster cold start than MLflow.

Open-source, self-hostable AI cost & workflow observability. Find the prompt, customer, model, and workflow path behind every LLM cost spike — without a proxy.

AgentGlass — a glass box for your AI agents. Capture, inspect, replay, and analyze every agent run (LLM observability & tracing). No SDK, no code changes.

Site Sleuth is a chatbot that assists users in efficiently using websites by analyzing and answering questions provided through a link.

A local LLM proxy tool that intercepts and visualizes all OpenAI API calls, providing a real-time web monitoring dashboard.

Open-source evaluation and observability platform for AI agent applications.

The comprehensive list of LLM observability tools: 120 open-source and commercial platforms for AI tracing, monitoring, evaluation, analytics, cost, and security.

Agent cognitive middleware — inspect, detect, and verify agent reasoning in real time. By LatticeAG.

The official JavaScript/TypeScript client library for sending AI application observability data to the AgentInsight platform.

Drop-in observability for LLM apps — quality scoring, hallucination detection, cost tracking and agent debugging

An experimental lab for studying how AI models collaborate and how their contributions evolve. Agents from different providers draft, refine and vote on plans; every call is traced, and the reports show what each one took, kept or lost.

Open-source cryptographic audit trail for AI agents. Produces cryptographic receipts - SHA-256 hash-chained, tamper-evident, verifiable offline. Supports EU AI Act Article 12 traceability.