Agent-triage – diagnosis of agent failures from production traces
Replays agent traces step-by-step to pinpoint exact failure turns automatically.
A Go-native framework for LLM agents, with OpenTelemetry observability built in.
One binary with embedded dashboard beats LangSmith's closed-source SaaS for Go teams.
Go developers building AI agents
LangChain · Eino · Genkit
Replays agent traces step-by-step to pinpoint exact failure turns automatically.
Replay, fork, diff, eval agent traces locally—like Git for agent behavior, fills a real gap.
Agent-native eval workflow beats LangSmith's manual dashboard setup.
Self-healing agents patch prompts automatically via replay validation; beats manual iteration.
Iteratively improves agent harnesses from 67% to 87% on tau-bench using production traces.
Agents write Python to analyze traces; 2x improvement on τ2-bench, but narrow evaluation scope.