Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Architecture audit that maps module dependencies, checks layering integrity, and flags structural decay across a codebase, drawing on twelve classic engineering books. Triggers when: user asks to audit architecture, review folder/module structure, check for circular imports, understand...
.claude/skills/sickn33-brooks-audit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-05 | ✓→✗ | ▼ Worse | -65% | 0% |
| case-04 | ✓→✗ | ▼ Worse | -42% | 0% |
| case-06 | ✓→✗ | ▼ Worse | -24% | 0% |
| case-07 | ✓→✗ | ▼ Worse | -29% | 0% |
Use this skill when you need architecture audit that maps module dependencies, checks layering integrity, and flags structural decay across a codebase, drawing on twelve classic engineering books. Triggers when: user asks to audit architecture, review folder/module structure, check for circular imports, understand...
../_shared/common.md for the Iron Law, Project Config, Report Template, and Health Score rules../_shared/source-coverage.md for book-level coverage, exceptions, and tradeoffs../_shared/decay-risks.md for symptom definitions and source attributionsarchitecture-guide.md in this directory for the audit frameworkOnboarding mode: If the user asks for an onboarding report, codebase tour, or "explain this codebase to a new developer", read onboarding-guide.md from this directory and follow it instead of architecture-guide.md. This mode explains rather than diagnoses — no Health Score, no Iron Law findings.
If the user has not specified files or a directory to audit: apply Auto Scope Detection from ../_shared/common.md to determine the audit scope before proceeding.
Mode line in report: Architecture Audit
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→fail | 14,660 | 3,745 | -74% | 1 | 1 | 0% | 2,576 | 907 | -65% | 0 | 0 | — |
case-11 | fail→fail | 4,445 | 4,895 | +10% | 1 | 1 | 0% | 698 | 718 | +3% | 0 | 0 | — |
case-01 | fail→fail | 13,287 | 5,131 | -61% | 1 | 1 | 0% | 1,965 | 716 | -64% | 0 | 0 | — |
case-02 | fail→fail | 11,195 | 4,112 | -63% | 1 | 1 | 0% | 1,762 | 714 | -59% | 0 | 0 | — |
case-03 | pass→pass | 15,440 | 18,679 | +21% | 1 | 1 | 0% | 2,566 | 3,448 | +34% | 0 | 0 | — |
case-04 | pass→fail | 7,847 | 4,222 | -46% | 1 | 1 | 0% | 1,222 | 707 | -42% | 0 | 0 | — |
case-06 | pass→fail | 6,803 | 4,299 | -37% | 1 | 1 | 0% | 1,047 | 794 | -24% | 0 | 0 | — |
case-07 | pass→fail | 6,512 | 5,693 | -13% | 1 | 1 | 0% | 1,016 | 724 | -29% | 0 | 0 | — |
case-08 | pass→fail | 16,091 | 4,626 | -71% | 1 | 1 | 0% | 2,665 | 681 | -74% | 0 | 0 | — |
case-09 | pass→fail | 10,192 | 3,580 | -65% | 1 | 1 | 0% | 1,945 | 753 | -61% | 0 | 0 | — |
case-10 | pass→fail | 14,582 | 4,630 | -68% | 1 | 1 | 0% | 2,466 | 749 | -70% | 0 | 0 | — |
case-12 | fail→pass | 16,660 | 4,241 | -75% | 1 | 1 | 0% | 2,705 | 1,110 | -59% | 0 | 0 | — |
case-13 | fail→fail | 10,306 | 4,688 | -55% | 1 | 1 | 0% | 1,751 | 774 | -56% | 0 | 0 | — |
case-14 | fail→fail | 12,624 | 4,080 | -68% | 1 | 1 | 0% | 1,876 | 740 | -61% | 0 | 0 | — |
case-15 | pass→fail | 13,672 | 3,812 | -72% | 1 | 1 | 0% | 2,114 | 641 | -70% | 0 | 0 | — |
case-16 | fail→fail | 4,010 | 3,463 | -14% | 1 | 1 | 0% | 666 | 755 | +13% | 0 | 0 | — |
case-17 | pass→pass | 12,681 | 20,110 | +59% | 1 | 1 | 0% | 2,217 | 1,774 | -20% | 0 | 0 | — |
case-18 | pass→pass | 7,444 | 10,135 | +36% | 1 | 1 | 0% | 1,407 | 2,455 | +74% | 0 | 0 | — |
case-19 | pass→pass | 12,934 | 11,292 | -13% | 1 | 1 | 0% | 2,594 | 2,816 | +9% | 0 | 0 | — |
case-20 | pass→pass | 15,195 | 19,485 | +28% | 1 | 1 | 0% | 2,650 | 4,353 | +64% | 0 | 0 | — |
case-21 | pass→pass | 10,515 | 11,072 | +5% | 1 | 1 | 0% | 2,059 | 2,624 | +27% | 0 | 0 | — |
case-22 | pass→pass | 10,423 | 5,863 | -44% | 1 | 1 | 0% | 1,880 | 1,592 | -15% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 9 counted toward the lift figure. The other 13 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -32 percentage points is the difference between those two pass rates over the 9 comparable cases. 8 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.