Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →athola/bug-review
Hunts bugs with evidence trails.
Passed 18 of 23 test cases with the skill, and 18 of 23 without it.
1 fixed19 passed either way4 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
A test case the skill fixed. Same model, same prompt, with and without the skill.
We need a full bug audit on our authentication middleware before going live. Please review the codebase for runtime errors, API misuse, and concurrency flaws. Structure your output into clear sections: Summary of scope, Defects Found (listing file:line reference, title, severity, anchor text, root cause, impact, and fix), Proposed Fixes with code diffs, Test Updates validating the remedies, and Evidence including commands executed and terminal logs.
| Checked | Without skill | With skill |
|---|---|---|
| Provides file and line locations for all identified authentication defects. | Missed | Passed |
| Provides proposed fixes formatted as standard code diffs. | Missed | Passed |
| Includes raw terminal output logs from execution commands. | Missed | Passed |
The instructions your model reads when the skill loads.
Systematic bug identification and fixing with language-specific expertise.
/bug-review
Verification: Run the command with --help flag to verify availability.
bug-review:language-detectedbug-review:repro-planbug-review:defects-documentedbug-review:fixes-preparedbug-review:verification-planbug-review:findings-verifiedLoad additional context as needed:
@include modules/language-detection.md - Manifest heuristics, expertise framing, version constraints@include modules/defect-documentation.md - Severity classification, root cause analysis, static analyzers@include modules/fix-preparation.md - Minimal patches, idiomatic patterns, test coveragetest cases passed, vs 18 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.