Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Language-specific auto-lint/format/typecheck pipeline. Supports Python (ruff+pyright), TypeScript (prettier+eslint+tsc), Go (gofmt+golangci-lint). Auto-fix and convergence loops.
.claude/skills/a5c-ai-quality-hooks/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 10% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -24% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -32% | 0% |
When quality checks fail:
Composite score = weighted average:
Target: 85/100 (configurable via targetQuality)
| Pilot Shell Hook | Quality Action | |-----------------|----------------| | PostToolUse (Edit/Write) | Run file_checker on modified files | | PostToolUse (file creation) | Run full pipeline on new files | | Pre-merge | Run full pipeline on all changes |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 13,115 | 11,350 | -13% | 1 | 1 | 0% | 2,074 | 2,277 | +10% | 0 | 0 | — |
case-02 | fail→pass | 14,682 | 8,189 | -44% | 1 | 1 | 0% | 2,416 | 1,746 | -28% | 0 | 0 | — |
case-03 | fail→pass | 17,057 | 11,888 | -30% | 1 | 1 | 0% | 3,100 | 2,370 | -24% | 0 | 0 | — |
case-04 | fail→fail | 11,135 | 5,024 | -55% | 1 | 1 | 0% | 1,825 | 980 | -46% | 0 | 0 | — |
case-05 | fail→pass | 13,667 | 9,813 | -28% | 1 | 1 | 0% | 2,332 | 1,943 | -17% | 0 | 0 | — |
case-06 | fail→pass | 4,760 | 1,851 | -61% | 1 | 1 | 0% | 753 | 512 | -32% | 0 | 0 | — |
case-07 | fail→pass | 11,436 | 1,901 | -83% | 1 | 1 | 0% | 1,835 | 465 | -75% | 0 | 0 | — |
case-08 | fail→pass | 7,496 | 1,573 | -79% | 1 | 1 | 0% | 1,228 | 418 | -66% | 0 | 0 | — |
case-09 | fail→pass | 9,621 | 1,389 | -86% | 1 | 1 | 0% | 1,534 | 403 | -74% | 0 | 0 | — |
case-10 | fail→pass | 9,367 | 1,886 | -80% | 1 | 1 | 0% | 1,428 | 435 | -70% | 0 | 0 | — |
case-11 | fail→pass | 9,085 | 3,241 | -64% | 1 | 1 | 0% | 1,236 | 504 | -59% | 0 | 0 | — |
case-12 | fail→pass | 10,954 | 2,288 | -79% | 1 | 1 | 0% | 1,933 | 553 | -71% | 0 | 0 | — |
case-13 | fail→fail | 9,788 | 2,006 | -80% | 1 | 1 | 0% | 1,784 | 516 | -71% | 0 | 0 | — |
case-14 | fail→fail | 10,205 | 4,687 | -54% | 1 | 1 | 0% | 1,631 | 924 | -43% | 0 | 0 | — |
case-15 | fail→fail | 14,431 | 9,046 | -37% | 1 | 1 | 0% | 2,805 | 2,145 | -24% | 0 | 0 | — |
case-16 | fail→pass | 10,296 | 1,657 | -84% | 1 | 1 | 0% | 1,805 | 449 | -75% | 0 | 0 | — |
case-17 | fail→pass | 15,819 | 3,563 | -77% | 1 | 1 | 0% | 2,357 | 786 | -67% | 0 | 0 | — |
case-18 | fail→fail | 4,579 | 1,441 | -69% | 1 | 1 | 0% | 719 | 396 | -45% | 0 | 0 | — |
case-19 | fail→pass | 9,296 | 1,790 | -81% | 1 | 1 | 0% | 1,682 | 498 | -70% | 0 | 0 | — |
case-20 | fail→fail | 10,498 | 10,119 | -4% | 1 | 1 | 0% | 2,094 | 2,328 | +11% | 0 | 0 | — |
case-21 | fail→fail | 7,410 | 8,277 | +12% | 1 | 1 | 0% | 1,436 | 1,795 | +25% | 0 | 0 | — |
case-22 | fail→fail | 6,472 | 6,047 | -7% | 1 | 1 | 0% | 1,154 | 1,383 | +20% | 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. The headline lift of +64 percentage points is the difference between those two pass rates over the 22 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.