Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Audit structural debt, logic leakage, and monolithic components across Web, Mobile, and Backend codebases. Use when reviewing architecture, assessing tech debt, detecting logic in wrong layers, or identifying God classes.
.claude/skills/hoangnguyen0403-common-architecture-audit/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -11% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -24% | 0% |
Identify split sources of truth by searching for redundant directory patterns.
Service.ts vs ServiceNew.ts vs ServiceV2.ts./v1, /v2 or "Refactor" folders.See implementation examples for detection scripts.
Find business logic trapped in wrong layer.
grep -rE "useEffect|useState|useMemo" components --include="*.tsx" | wc -l — if components/ hook count > 20x hooks/ folder, architecture monolithic.grep -rE "http\.|dio\.|socket\." lib/widgets --include="*.dart" | wc -l — I/O or state mutation > 5 lines in build() High Debt.grep -rE "Repository\.|Query\.|db\." src/controllers --include="*.ts" | wc -l — Controllers must only handle request parsing and response formatting.Flag massive files violating Single Responsibility Principle.
See implementation examples for monolith detection scripts.
Check for large metadata or constants impacting IDE performance and binary size.
See implementation examples for resource audit scripts.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 28,890 | 47,152 | +63% | 1 | 1 | 0% | 4,607 | 4,242 | -8% | 0 | 0 | — |
case-02 | fail→pass | 24,261 | 32,839 | +35% | 1 | 1 | 0% | 3,665 | 4,734 | +29% | 0 | 0 | — |
case-03 | fail→pass | 28,742 | 43,260 | +51% | 1 | 1 | 0% | 4,619 | 4,133 | -11% | 0 | 0 | — |
case-04 | fail→pass | 25,492 | 13,495 | -47% | 1 | 1 | 0% | 2,584 | 2,729 | +6% | 0 | 0 | — |
case-05 | fail→pass | 24,587 | 8,061 | -67% | 1 | 1 | 0% | 2,315 | 1,766 | -24% | 0 | 0 | — |
case-06 | fail→pass | 16,783 | 14,953 | -11% | 1 | 1 | 0% | 2,663 | 2,736 | +3% | 0 | 0 | — |
case-07 | fail→pass | 15,541 | 14,306 | -8% | 1 | 1 | 0% | 2,680 | 2,378 | -11% | 0 | 0 | — |
case-08 | fail→pass | 41,632 | 14,705 | -65% | 1 | 1 | 0% | 2,604 | 1,131 | -57% | 0 | 0 | — |
case-09 | pass→pass | 15,194 | 8,980 | -41% | 1 | 1 | 0% | 2,387 | 1,778 | -26% | 0 | 0 | — |
case-10 | fail→pass | 15,024 | 3,288 | -78% | 1 | 1 | 0% | 2,134 | 986 | -54% | 0 | 0 | — |
case-11 | fail→pass | 13,584 | 2,508 | -82% | 1 | 1 | 0% | 2,174 | 807 | -63% | 0 | 0 | — |
case-12 | fail→pass | 25,527 | 2,850 | -89% | 1 | 1 | 0% | 1,929 | 874 | -55% | 0 | 0 | — |
case-13 | pass→pass | 18,370 | 15,559 | -15% | 1 | 1 | 0% | 2,933 | 2,840 | -3% | 0 | 0 | — |
case-14 | fail→fail | 21,719 | 19,643 | -10% | 1 | 1 | 0% | 3,059 | 3,916 | +28% | 0 | 0 | — |
case-15 | pass→pass | 27,254 | 31,597 | +16% | 1 | 1 | 0% | 2,992 | 3,384 | +13% | 0 | 0 | — |
case-16 | fail→pass | 15,412 | 6,283 | -59% | 1 | 1 | 0% | 2,399 | 1,309 | -45% | 0 | 0 | — |
case-17 | fail→fail | 26,511 | 12,983 | -51% | 1 | 1 | 0% | 2,418 | 2,400 | -1% | 0 | 0 | — |
case-18 | fail→fail | 11,470 | 5,995 | -48% | 1 | 1 | 0% | 1,812 | 1,019 | -44% | 0 | 0 | — |
case-19 | pass→pass | 16,364 | 9,310 | -43% | 1 | 1 | 0% | 2,456 | 1,673 | -32% | 0 | 0 | — |
case-20 | pass→pass | 25,743 | 8,341 | -68% | 1 | 1 | 0% | 1,891 | 1,784 | -6% | 0 | 0 | — |
case-21 | fail→pass | 17,347 | 18,450 | +6% | 1 | 1 | 0% | 2,357 | 2,233 | -5% | 0 | 0 | — |
case-22 | fail→pass | 25,788 | 2,283 | -91% | 1 | 1 | 0% | 1,956 | 736 | -62% | 0 | 0 | — |
case-23 | pass→pass | 11,706 | 55,597 | +375% | 1 | 1 | 0% | 1,607 | 4,686 | +192% | 0 | 0 | — |
case-24 | pass→pass | 21,507 | 22,630 | +5% | 1 | 1 | 0% | 4,159 | 5,325 | +28% | 0 | 0 | — |
case-25 | pass→pass | 8,051 | 19,076 | +137% | 1 | 1 | 0% | 1,213 | 3,528 | +191% | 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. 25 cases were attempted. The headline lift of +56 percentage points is the difference between those two pass rates over the 25 comparable cases.
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.