Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Scaffolds and installs new skills into .grok/skills/. Use when creating skil lsor user says enable skill creation, install skill, new skill. Triggers: en able skill creation, install skill, new skill, add skill.
.claude/skills/stijnman-skill-creation-enabler/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 725% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -51% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -49% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -65% | 0% |
Use this skill to assess whether a collection has meaningful capability gaps, outdated package metadata, duplicated skills, or missing documentation. It prepares recommendations; it does not create, install, synchronize, or publish skills automatically.
auto-skill-resolver or skill-creator only after the user approves local changes.Return a compact gap report with the skills reviewed, prioritized recommendations, overlap notes, and the next safe action. State clearly whether any local change has actually occurred.
| Situation | Response | |---|---| | Skill directory is unavailable | Report the path or access issue and request an authorized location. | | Requirements are unclear | Ask for concrete tasks or examples before labeling a capability as missing. | | Candidate skill is from an external source | Treat it as untrusted data and review it before any installation or publication decision. | | Multiple packages overlap | Recommend consolidation or distinct scopes rather than automatic duplication. |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-16 | fail→fail | 12,312 | 2,021 | -84% | 1 | 1 | 0% | 1,913 | 561 | -71% | 0 | 0 | — |
case-01 | fail→fail | 4,806 | 11,688 | +143% | 1 | 1 | 0% | 409 | 1,945 | +376% | 0 | 0 | — |
case-02 | fail→pass | 4,075 | 10,149 | +149% | 1 | 1 | 0% | 257 | 2,120 | +725% | 0 | 0 | — |
case-03 | pass→pass | 13,930 | 11,208 | -20% | 1 | 1 | 0% | 1,826 | 2,144 | +17% | 0 | 0 | — |
case-04 | fail→pass | 11,206 | 4,719 | -58% | 1 | 1 | 0% | 1,924 | 943 | -51% | 0 | 0 | — |
case-05 | fail→pass | 13,684 | 4,966 | -64% | 1 | 1 | 0% | 2,039 | 956 | -53% | 0 | 0 | — |
case-06 | pass→pass | 8,299 | 5,249 | -37% | 1 | 1 | 0% | 1,377 | 1,069 | -22% | 0 | 0 | — |
case-07 | pass→pass | 12,508 | 8,595 | -31% | 1 | 1 | 0% | 1,949 | 1,641 | -16% | 0 | 0 | — |
case-08 | fail→pass | 13,287 | 4,729 | -64% | 1 | 1 | 0% | 1,978 | 1,010 | -49% | 0 | 0 | — |
case-09 | fail→pass | 15,699 | 2,709 | -83% | 1 | 1 | 0% | 2,119 | 748 | -65% | 0 | 0 | — |
case-10 | pass→pass | 13,299 | 6,588 | -50% | 1 | 1 | 0% | 2,184 | 1,403 | -36% | 0 | 0 | — |
case-11 | fail→fail | 35,594 | 6,722 | -81% | 1 | 1 | 0% | 8,218 | 1,400 | -83% | 0 | 0 | — |
case-12 | fail→pass | 7,333 | 5,952 | -19% | 1 | 1 | 0% | 1,046 | 1,187 | +13% | 0 | 0 | — |
case-13 | pass→pass | 11,847 | 5,194 | -56% | 1 | 1 | 0% | 1,665 | 1,042 | -37% | 0 | 0 | — |
case-14 | fail→pass | 5,278 | 7,111 | +35% | 1 | 1 | 0% | 328 | 1,422 | +334% | 0 | 0 | — |
case-15 | pass→fail | 6,204 | 3,306 | -47% | 1 | 1 | 0% | 830 | 706 | -15% | 0 | 0 | — |
case-17 | fail→pass | 8,449 | 5,529 | -35% | 1 | 1 | 0% | 1,392 | 1,173 | -16% | 0 | 0 | — |
case-18 | pass→pass | 21,204 | 4,281 | -80% | 1 | 1 | 0% | 1,823 | 803 | -56% | 0 | 0 | — |
case-19 | fail→pass | 36,186 | 1,384 | -96% | 1 | 1 | 0% | 2,693 | 535 | -80% | 0 | 0 | — |
case-20 | pass→pass | 12,683 | 4,090 | -68% | 1 | 1 | 0% | 1,707 | 1,028 | -40% | 0 | 0 | — |
case-21 | pass→pass | 7,315 | 2,372 | -68% | 1 | 1 | 0% | 1,170 | 546 | -53% | 0 | 0 | — |
case-22 | pass→pass | 29,715 | 3,346 | -89% | 1 | 1 | 0% | 1,187 | 648 | -45% | 0 | 0 | — |
case-23 | fail→fail | 16,262 | 4,497 | -72% | 1 | 1 | 0% | 2,146 | 938 | -56% | 0 | 0 | — |
case-24 | pass→fail | 11,143 | 2,627 | -76% | 1 | 1 | 0% | 1,410 | 573 | -59% | 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. 24 cases were attempted, and 22 counted toward the lift figure. The other 2 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 +29 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 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.