Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Extract durable working preferences from recent Cursor chats and convert them into skills, rules, or workflow docs. Use when asked to learn preferences, mine feedback, personalize workflows, or generate team/person-specific agent guidance.
.claude/skills/kunanonj-cursor-plugin-teamkit-workflow-from-chats/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 408% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 508% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 160% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -30% | 0% |
Infer durable working preferences from recent chats. Do not summarize chats; extract reusable workflow guidance.
Return a concise synthesis first:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 6,300 | 27,498 | +336% | 1 | 1 | 0% | 1,058 | 5,376 | +408% | 0 | 0 | — |
case-02 | pass→pass | 10,576 | 19,237 | +82% | 1 | 1 | 0% | 1,709 | 4,047 | +137% | 0 | 0 | — |
case-03 | pass→pass | 9,940 | 12,806 | +29% | 1 | 1 | 0% | 1,583 | 2,707 | +71% | 0 | 0 | — |
case-04 | pass→pass | 7,113 | 10,977 | +54% | 1 | 1 | 0% | 1,134 | 2,469 | +118% | 0 | 0 | — |
case-05 | pass→pass | 10,251 | 9,353 | -9% | 1 | 1 | 0% | 1,622 | 2,060 | +27% | 0 | 0 | — |
case-06 | pass→pass | 9,086 | 13,195 | +45% | 1 | 1 | 0% | 1,597 | 2,956 | +85% | 0 | 0 | — |
case-07 | fail→pass | 2,944 | 12,755 | +333% | 1 | 1 | 0% | 440 | 2,677 | +508% | 0 | 0 | — |
case-08 | fail→pass | 7,953 | 19,185 | +141% | 1 | 1 | 0% | 1,225 | 3,187 | +160% | 0 | 0 | — |
case-09 | pass→pass | 13,347 | 5,560 | -58% | 1 | 1 | 0% | 2,028 | 1,344 | -34% | 0 | 0 | — |
case-10 | fail→pass | 10,018 | 12,186 | +22% | 1 | 1 | 0% | 1,885 | 2,633 | +40% | 0 | 0 | — |
case-11 | pass→pass | 3,732 | 2,633 | -29% | 1 | 1 | 0% | 612 | 960 | +57% | 0 | 0 | — |
case-12 | pass→pass | 5,929 | 2,737 | -54% | 1 | 1 | 0% | 918 | 979 | +7% | 0 | 0 | — |
case-13 | pass→pass | 11,239 | 6,491 | -42% | 1 | 1 | 0% | 1,897 | 1,624 | -14% | 0 | 0 | — |
case-14 | pass→pass | 8,524 | 3,422 | -60% | 1 | 1 | 0% | 1,465 | 978 | -33% | 0 | 0 | — |
case-15 | pass→pass | 5,818 | 2,545 | -56% | 1 | 1 | 0% | 971 | 871 | -10% | 0 | 0 | — |
case-16 | fail→pass | 8,412 | 2,594 | -69% | 1 | 1 | 0% | 1,253 | 877 | -30% | 0 | 0 | — |
case-17 | pass→pass | 8,248 | 7,477 | -9% | 1 | 1 | 0% | 1,392 | 1,607 | +15% | 0 | 0 | — |
case-18 | pass→pass | 9,320 | 7,741 | -17% | 1 | 1 | 0% | 1,559 | 1,862 | +19% | 0 | 0 | — |
case-19 | fail→pass | 8,581 | 5,803 | -32% | 1 | 1 | 0% | 1,452 | 1,408 | -3% | 0 | 0 | — |
case-20 | fail→pass | 13,556 | 3,001 | -78% | 1 | 1 | 0% | 2,388 | 958 | -60% | 0 | 0 | — |
case-21 | pass→pass | 6,798 | 5,332 | -22% | 1 | 1 | 0% | 1,128 | 1,308 | +16% | 0 | 0 | — |
case-22 | fail→pass | 10,525 | 4,119 | -61% | 1 | 1 | 0% | 1,658 | 1,120 | -32% | 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.