Paths/Claude Engineer/Testing with Claude Code / Write the Test
Ask for the Failing Test First
A test that fails first proves it can fail at all. Write the brief that asks for one.
Validator
IDLE — checks run live as you type commands; submit when done
01Requirements
live checks- Directory ~/.claude/testing exists○
- File ~/.claude/testing/brief.md exists○
- The brief states the behaviour and the case○
- The brief states that the test must be seen failing○
02Walkthrough
4 steps- 01
Create the testing directory
The briefs live with the rest of the Claude configuration.
mkdir -p ~/.claude/testing - 02
Write the behaviour and the case
One behaviour, one case. The agent writes the test from the behaviour, not from the code it will later change.
echo '# Test brief\n\n## The behaviour\n\nThe refund endpoint refuses a refund larger than the captured amount.\n\n## The case\n\nA refund of 120 on a capture of 100.' > ~/.claude/testing/brief.md - 03
Write the ordering rule
Write the test, run it and watch it fail, and only then write the implementation. The failure is the proof the test works.
echo '\n## The rule\n\nWrite the test before the implementation, then run it and watch it fail. The failing output is the proof that the test is testing something.' >> ~/.claude/testing/brief.md - 04
Verify and submit
cat the brief, then submit for validation.
cat ~/.claude/testing/brief.md
03Stuck?
0 of 3 revealedTake a hint without losing the lab — each one points at the thing to look at, not at the answer.
04AI Tutor
guides, never solvesTell me what you're stuck on — I'll nudge, not solve. Which requirement are you looking at?
04Notebook
saved locally in this browserlearner@lab — ubuntu:24.04
Session is disposable
↑/↓ recalls commands · Ctrl+Enter submits · type help for the command list