Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Deep audit of a skills directory against the Skill Creator standard. Produces a scored report and phased remediation plan.
.claude/skills/hoangnguyen0403-battle-test/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -54% | 0% |
> !IMPORTANT] > Deep audit of a skills directory against the Skill Creator standard. Produces a scored report and phased remediation plan.
Optional args: slug=<feature>, ticket=<id/url>, mode=interactive|autonomous|channel, channel=<id>, auto_continue=true|false, profile=business|hybrid|technical.
When the user asks to perform this workflow, execute the following steps:
> Goal: Evaluate every SKILL.md in the target directory against common-skill-creator. Deliver a quantified health report and prioritized remediation plan.
Identify the tech stack and all skill files.
bash# Count total skills per category find . -name "SKILL.md" | sed 's|/[^/]*/SKILL.md||' | sort | uniq -c
Run scans to detect format and structure violations.
grep -rL "triggers:\|priority:\|Anti-Patterns" <SKILLS>/grep -r "src/\*\*" <SKILLS>/find . -name "SKILL.md" -exec awk 'END{if(NR>100) print FILENAME": "NR" lines"}' {} \;
Pick every P0 (CRITICAL) and a random sample of P1/P2 skills. Evaluate them against the Grading Rubric in: <SKILLS>/common/common-skill-creator/references/rubric.md when synced.
**No X**: Do Y. anti-patterns?Scoring Algorithm: Start at 100 points for each category. Apply deductions for findings (🔴-15 / 🟠-8 / 🟡-3 / 🔵-1).
Output the report using the Battle Test Report and Phased Plan templates in: <SKILLS>/common/common-skill-creator/references/rubric.md when synced.
task.md for Phase 1 remediation?"security)?"| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-08 | fail→fail | 15,573 | 10,636 | -32% | 1 | 1 | 0% | 2,549 | 2,439 | -4% | 0 | 0 | — |
case-01 | fail→fail | 12,585 | 4,571 | -64% | 1 | 1 | 0% | 2,299 | 839 | -64% | 0 | 0 | — |
case-02 | fail→fail | 27,051 | 5,235 | -81% | 1 | 1 | 0% | 5,118 | 1,008 | -80% | 0 | 0 | — |
case-18 | fail→pass | 13,758 | 2,445 | -82% | 1 | 1 | 0% | 2,198 | 1,017 | -54% | 0 | 0 | — |
case-03 | fail→fail | 15,775 | 5,549 | -65% | 1 | 1 | 0% | 2,601 | 995 | -62% | 0 | 0 | — |
case-04 | fail→pass | 11,349 | 1,393 | -88% | 1 | 1 | 0% | 1,285 | 840 | -35% | 0 | 0 | — |
case-05 | fail→pass | 10,521 | 5,211 | -50% | 1 | 1 | 0% | 1,970 | 1,647 | -16% | 0 | 0 | — |
case-06 | fail→fail | 7,629 | 4,763 | -38% | 1 | 1 | 0% | 1,320 | 1,234 | -7% | 0 | 0 | — |
case-07 | fail→pass | 4,427 | 5,676 | +28% | 1 | 1 | 0% | 681 | 1,218 | +79% | 0 | 0 | — |
case-09 | pass→pass | 8,182 | 1,954 | -76% | 1 | 1 | 0% | 1,446 | 956 | -34% | 0 | 0 | — |
case-10 | fail→pass | 11,786 | 2,662 | -77% | 1 | 1 | 0% | 2,029 | 942 | -54% | 0 | 0 | — |
case-11 | fail→pass | 8,797 | 1,930 | -78% | 1 | 1 | 0% | 1,517 | 935 | -38% | 0 | 0 | — |
case-12 | fail→pass | 11,144 | 1,937 | -83% | 1 | 1 | 0% | 1,786 | 940 | -47% | 0 | 0 | — |
case-13 | fail→pass | 7,988 | 1,379 | -83% | 1 | 1 | 0% | 1,087 | 816 | -25% | 0 | 0 | — |
case-14 | fail→pass | 7,309 | 2,501 | -66% | 1 | 1 | 0% | 1,191 | 1,050 | -12% | 0 | 0 | — |
case-15 | fail→pass | 10,028 | 1,541 | -85% | 1 | 1 | 0% | 1,599 | 837 | -48% | 0 | 0 | — |
case-16 | pass→pass | 9,849 | 4,691 | -52% | 1 | 1 | 0% | 1,390 | 1,376 | -1% | 0 | 0 | — |
case-17 | pass→pass | 12,660 | 6,588 | -48% | 1 | 1 | 0% | 2,091 | 1,715 | -18% | 0 | 0 | — |
case-19 | fail→pass | 6,796 | 2,408 | -65% | 1 | 1 | 0% | 1,083 | 975 | -10% | 0 | 0 | — |
case-20 | pass→pass | 17,234 | 17,371 | +1% | 1 | 1 | 0% | 3,606 | 4,054 | +12% | 0 | 0 | — |
case-21 | pass→pass | 9,740 | 6,809 | -30% | 1 | 1 | 0% | 1,776 | 1,797 | +1% | 0 | 0 | — |
case-22 | pass→pass | 14,606 | 9,997 | -32% | 1 | 1 | 0% | 2,642 | 2,285 | -14% | 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 19 counted toward the lift figure. The other 3 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 +50 percentage points is the difference between those two pass rates over the 19 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.