When an agent says a task is done, what do you need to see before you accept it?
retained text
Founder disclosure: I build HyperHat, a baseline for how well teams direct AI coding agents. What I keep reading in threads here is that the agent's own summary is the least useful evidence. The tests passing is better, but a few threads this month describe tests that were mocked around the thing they were meant to check. So I am curious what actually moves you from "the agent says it is done" to "I will merge this": - Is it a specific artifact, like a diff against the original request, a test you wrote yourself, or a second reviewer? - Does it change when the person reviewing did not write the request? - Has anything you trusted turned out to be wrong in a way that changed what you ask for now? I am asking to learn, not to sell anything.
document fetched 2026-11-19 00:41 · canonical https://reddit.com/r/ClaudeCode/comments/1xzpomn/when_an_agent_says_a_task_is_done_what_do_you