Audit tests
Agent skills for agentic coding tools. Extremely opinionated. Updated (almost) daily.
npx -y skills add JHostalek/dotclaude --skill audit-testsAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 10 stars10 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use when the user wants a thorough audit of the existing test suite — aggressively cutting tests that exist just to exist, rewriting weak ones to actually catch the bugs they should, identifying production code that is uncovered or thinly covered, and adding tests where coverage would catch real defects. Trigger on "audit tests", "find test bloat", "tests don't catch anything", or similar requests.
SKILL.md
1.8 KB, as published. Nobody here has run it
!cat "${CLAUDE_SKILL_DIR}/../audit-workflow.md"
Run as the tests dimension. Lens:
A test earns its keep only if it catches context loss between changes or encodes domain knowledge the code can't express; doubt → delete, and deletion is auto-fix. Bloat to cut: restated-behavior, getter/setter coverage, mock-the-world ceremonies, framework smoke checks, "returns what I told it to return", assertions mirroring implementation line-for-line.
Rewrite weak tests to catch the bug they nominally guard: tests passing against a broken implementation, asserting incidental output over contract, or so heavily mocked the unit under test never runs. Rewrite preserving intended behavior = auto-fix; rewrite changing which behavior is correct = sign-off.
Add tests where coverage would catch real defects in uncovered or thinly covered production code — critical paths (auth, money, data integrity), non-obvious edges, business rules, integration seams. Skip framework behavior and passthroughs. Prefer integration over mocked unit, behavior over implementation. Coverage for existing behavior = auto-fix.
Drop: absent tests for trivial passthroughs, missing unit test where an integration test already exercises the path, coverage gaps in throwaway/generated code. Green baseline before cutting = engine's concern, not a finding.