AgentOps

The observability and DevTool platform for AI agents: session replays, cost and token tracking, and failure analysis via a lightweight SDK spanning the major agent frameworks.

View source on GitHub

Key takeaways

• 01

  Session replay is the debugging primitive for agent runs

• 02

  Track cost per session from day one - it shapes design

Flows built on this research

Harness Engineering

Observability for Agent Runs

Instrument your agent like production software: traces, session replay, cost tracking, and failure analysis on open standards.

4 steps · 90-120 minutes

Agent QA & Security

Audit Trails for Tool Calls

Record an immutable, queryable trail of every action your agent takes - who, what, when, why, and outcome - for security and accountability.

4 steps · 60-90 minutes

Agent QA & Security

Incident Response Runbook for Agent Mishaps

Prepare for the day your agent does damage: detection, containment, rollback, investigation, and prevention - written before you need it.

4 steps · 60-90 minutes