agent-tracing

Local-first agent debugger with replay, failure memory, smart highlights, and drift detection.

Open-source, self-hosted AI gateway for multi-tenant orgs: OpenAI-compatible LLM routing, RAG & vector stores, MCP hub, GPU fleet (MIG slicing), sandboxed coding agents, AI red-teaming & guardrails, and cost optimization — all in one console.

Self-evolving multi-agent writing platform — trace-driven evolution loop with human-gated releases

OrcaReplay — Time travel for AI agents. Record, replay, fork, and debug any agent run with any model. Built by the OrcaRouter.ai team.

Shared context substrate for AI agents. Retrieval that learns what's useful. Runs local or cloud.

Capture and analyze Claude Code sessions locally to track every tool call, decision, and reasoning step without external dependencies.

We check whether your agent has an alibi for what it did.

看清 AI 编码智能体的每一次可观察行动。 · Trace your agents, down to every tool and MCP.

Local-first debugger and observability toolkit for AI coding agents. Trace tools, commands, files, errors, latency, tokens, and agent runs.

Observability and enforcement for AI agent harnesses. Capture every run and runtime reliability with policy enforcement.

Open-source evaluation framework for LLM agents. Run head-to-head A/B tests, score with LLM-as-judge rubrics, and visualize results in a Streamlit dashboard. Model-agnostic, self-hosted, zero external infrastructure.

🌊 WaveLength is a production-shaped multi-agent customer support system built with React, FastAPI & LangGraph. 🤖 It uses supervisor routing, specialist agents, verified customer context, persistent memory, database-backed tools, and live tracing to make every agent decision, tool call, and workflow step visible. 🚀

Comprehensive agent analytics suite for AI agents built with the Google Agent Development Kit (ADK) , LangChain or other popular frameworks, powered by BigQuery Agent Analytics plugin and Grafana

AI coding agent observability and causal tracing for Claude Code, Codex CLI, OpenCode, and Python workflows

Langfuse observability for DeepSeek Harness (dsh): exports agent sessions as OpenTelemetry trace trees (GenAI semconv) to Langfuse's OTLP endpoint

Drop-in agent observability for LangChain, LangGraph, OpenAI Agents, Claude Agent SDK, and any OpenTelemetry-instrumented agent — ship traces, tool calls, tokens, cost, and latency to Cognipeer Console in two lines of code.

Official TypeScript/JavaScript SDK for Cognipeer Console — OpenAI-compatible chat, batch, realtime voice, embeddings, RAG, MCP, agent tracing, and guardrails for multi-tenant AI products.

Inspect Codex and Claude Code sessions as timelines, sequence diagrams, and raw events.

subs is an agent harness for the cloud. It runs an unprivileged agent loop and uses MCP servers for executing tools. Use it locally, remotely and with your team. You can customize everything.

Hand-built LLM agent observability: trace logging, token budget/sliding window, and LLM-as-judge scoring — no external SaaS. Wraps any agent function with timing, context truncation, and semantic (not string-match) answer evaluation.