Testing Commands
A curated collection of 5 testing commands for building with AI coding agents.
Command
Fix Failing Test
Diagnose and fix a failing test by finding the real root cause.
/fix-failing-test[test name or path]
Command
Hunt Flaky Tests
Reproduce a flaky test, find the real source of nondeterminism, and fix the cause.
/flaky-test-hunt<suspected test or area (optional)>
Command
Generate E2E Test
Scaffold a resilient end-to-end test for a user flow grounded in the real UI.
/generate-e2e-test<user flow to test>
Command
Run Evals
Run the project's LLM evaluation suite (DeepEval, promptfoo, or RAGAS) and report scores against thresholds before a merge.
/run-evals<eval suite path / config, or the feature to evaluate>
Command
Write Tests
Generate tests covering the happy path and edge cases for the given target.
/write-tests[file or function]