🔬 Evals

Automatic checks that tell you if your agent is doing good work. Everything runs on your machine, on your own key.

⚖️ The judge

Session quality, last 24 hours

🔬 Evaluators on your agent

Named checks ClawMetry runs to tell you if your agent is doing good work. Each one answers a plain question.

🧾 Recently scored sessions

🏅 Golden test suites