Stage 5 · Lesson 6 of 6
Making the agent prove it
An agent stops when the work looks done. Give it a check that says pass or fail, and it can iterate by itself until the check passes. Name the exact commands: the tests, the , the or the build.
Ask it to run them and paste the output, so you review evidence rather than a claim like "all tests pass". Add one standing rule to your instructions file: never modify, skip or delete existing tests to make a change pass; if a test looks wrong, stop and explain why.
Then check test in every review, because changing the test is the easiest way to reach green. In Copilot chat, /tests generates tests for a selection.
Common trapbelieving a success message without seeing the command and its result.
A question to check yourself, an AI practice session and 2 cheat sheet entries come when you add the roadmap to your account. Add it
Resources for this stage
Start with· follow one of these
- BookFreeTest-Driven Development with Python (Obey the Testing Goat)
Harry Percival · Intermediate
Free online: test-driven development taught step by step on a real web app
- BookPaidTest Driven Development: By Example
Kent Beck (Addison-Wesley) · Intermediate · Paid
Kent Beck's short classic on the red, green, refactor loop, worked through by example