Implementing and Evaluating a Basic Per-Action Monitor for Safer Evals
Source
metr.org
Date
Why it matters
A real postmortem-style walkthrough of building a per-action AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → monitor, including failure modes like an agent self-approving blocked actions and monitors blind to subagents.
Terms in this piece · Glossary
LLM — A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
