Run the test suite, analyze results, and propose fixes for any failures.
Scope
$ARGUMENTSis empty or "all": runmake test(full suite, fail-fast by tier).$ARGUMENTSis "unit": runmake test-unit.$ARGUMENTSis "integration": runmake test-integration.$ARGUMENTSis "agent": runmake test-agent.$ARGUMENTSis a file path: runmake test-file FILE=<path>.
Process
- Run the specified tests.
- If all pass: report success with a brief summary.
- If any fail:
- Parse the error output to identify the root cause.
- Read the failing test and the code under test.
- Propose a fix — explain what went wrong and how to resolve it.
- Do NOT apply the fix automatically — present it for approval.
Output
Write detailed results to scratch/test_results_latest.md with:
- Pass/fail counts per tier
- Failure details (test name, error message, relevant code)
- Proposed fixes if applicable
Return a ≤5 line summary to the main context.
Converted and distributed by TomeVault — claim your Tome and manage your conversions.