Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Govern how the design system evolves — SemVer for tokens/components, the contribution workflow, deprecation policy, and change communication. Use when the user wants to add/promote/deprecate a component or token, decide a version bump, set up a contribution process, or keep the system from fragmenting.
.claude/skills/plugin87-governance/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 33% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -61% | 0% |
Keep the system consistent as it grows. Apply versioning, contribution, and deprecation rules.
workflows/governance.md (SemVer table, contribution workflow, deprecation policy, change comms).design-systems/crosswalk.md style); remove only in a major.CLAUDE.md (File Reference Map + relevant table/router) and add a changelog entry.python3 scripts/validate_tokens.py passes; contrast re-checked if colors changed.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | fail→pass | 14,921 | 14,754 | -1% | 1 | 1 | 0% | 2,274 | 2,393 | +5% | 0 | 0 | — |
case-07 | pass→pass | 4,080 | 3,562 | -13% | 1 | 1 | 0% | 644 | 806 | +25% | 0 | 0 | — |
case-01 | fail→pass | 17,211 | 9,911 | -42% | 1 | 1 | 0% | 1,949 | 2,005 | +3% | 0 | 0 | — |
case-03 | pass→pass | 9,653 | 8,481 | -12% | 1 | 1 | 0% | 1,478 | 1,893 | +28% | 0 | 0 | — |
case-04 | pass→pass | 8,446 | 6,487 | -23% | 1 | 1 | 0% | 1,560 | 1,468 | -6% | 0 | 0 | — |
case-05 | pass→pass | 7,751 | 5,816 | -25% | 1 | 1 | 0% | 1,298 | 1,156 | -11% | 0 | 0 | — |
case-06 | fail→fail | 9,324 | 6,385 | -32% | 1 | 1 | 0% | 1,458 | 1,256 | -14% | 0 | 0 | — |
case-08 | fail→fail | 5,161 | 4,606 | -11% | 1 | 1 | 0% | 779 | 817 | +5% | 0 | 0 | — |
case-09 | fail→fail | 10,547 | 5,307 | -50% | 1 | 1 | 0% | 1,655 | 1,204 | -27% | 0 | 0 | — |
case-10 | fail→pass | 8,824 | 6,824 | -23% | 1 | 1 | 0% | 1,297 | 1,322 | +2% | 0 | 0 | — |
case-11 | fail→pass | 6,656 | 5,890 | -12% | 1 | 1 | 0% | 967 | 1,286 | +33% | 0 | 0 | — |
case-12 | fail→pass | 10,099 | 1,864 | -82% | 1 | 1 | 0% | 1,497 | 578 | -61% | 0 | 0 | — |
case-13 | pass→pass | 6,218 | 2,709 | -56% | 1 | 1 | 0% | 963 | 727 | -25% | 0 | 0 | — |
case-14 | fail→pass | 14,616 | 10,603 | -27% | 1 | 1 | 0% | 2,389 | 1,937 | -19% | 0 | 0 | — |
case-15 | pass→pass | 4,683 | 3,153 | -33% | 1 | 1 | 0% | 687 | 784 | +14% | 0 | 0 | — |
case-16 | pass→pass | 10,713 | 6,305 | -41% | 1 | 1 | 0% | 1,536 | 1,215 | -21% | 0 | 0 | — |
case-17 | fail→pass | 5,765 | 3,845 | -33% | 1 | 1 | 0% | 882 | 938 | +6% | 0 | 0 | — |
case-18 | fail→pass | 9,994 | 6,789 | -32% | 1 | 1 | 0% | 1,580 | 1,360 | -14% | 0 | 0 | — |
case-19 | pass→pass | 16,289 | 5,998 | -63% | 1 | 1 | 0% | 2,570 | 1,212 | -53% | 0 | 0 | — |
case-20 | pass→pass | 11,110 | 7,305 | -34% | 1 | 1 | 0% | 1,620 | 1,349 | -17% | 0 | 0 | — |
case-21 | pass→pass | 9,913 | 3,261 | -67% | 1 | 1 | 0% | 1,519 | 797 | -48% | 0 | 0 | — |
case-22 | pass→pass | 4,549 | 2,344 | -48% | 1 | 1 | 0% | 693 | 682 | -2% | 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 +36 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.