Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Disable continuous learning mode
.claude/skills/aiskillstore-learn-off/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -79% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -81% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -72% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -67% | 0% |
Disable continuous learning mode. Automatic insight extraction will stop.
Deactivates continuous learning mode:
/learn commands still workknowledge/state.jsonjson { "learning_mode": false, "learning_mode_since": null }
knowledge/state.jsonContinuous Learning: DISABLED
─────────────────────────────
Learning mode is now inactive.
Session summary:
- Extractions performed: X
- Queries analyzed: Y
- Insights captured: Z
Manual extraction is still available via /learn.
Use /knowledge to view accumulated insights./learn-on at any time| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 4,500 | 7,249 | +61% | 1 | 1 | 0% | 609 | 849 | +39% | 0 | 0 | — |
case-02 | fail→fail | 5,156 | 3,445 | -33% | 1 | 1 | 0% | 702 | 867 | +24% | 0 | 0 | — |
case-03 | fail→fail | 5,171 | 6,667 | +29% | 1 | 1 | 0% | 723 | 737 | +2% | 0 | 0 | — |
case-04 | fail→pass | 6,110 | 1,568 | -74% | 1 | 1 | 0% | 938 | 445 | -53% | 0 | 0 | — |
case-05 | fail→pass | 15,864 | 1,787 | -89% | 1 | 1 | 0% | 2,470 | 517 | -79% | 0 | 0 | — |
case-06 | pass→pass | 9,883 | 2,235 | -77% | 1 | 1 | 0% | 1,538 | 547 | -64% | 0 | 0 | — |
case-07 | pass→pass | 7,786 | 2,646 | -66% | 1 | 1 | 0% | 1,150 | 628 | -45% | 0 | 0 | — |
case-08 | fail→pass | 13,518 | 1,727 | -87% | 1 | 1 | 0% | 2,327 | 449 | -81% | 0 | 0 | — |
case-09 | fail→pass | 9,722 | 1,626 | -83% | 1 | 1 | 0% | 1,577 | 438 | -72% | 0 | 0 | — |
case-10 | fail→pass | 8,810 | 1,722 | -80% | 1 | 1 | 0% | 1,283 | 421 | -67% | 0 | 0 | — |
case-11 | fail→pass | 11,448 | 2,376 | -79% | 1 | 1 | 0% | 1,605 | 539 | -66% | 0 | 0 | — |
case-12 | pass→pass | 9,242 | 2,782 | -70% | 1 | 1 | 0% | 1,321 | 688 | -48% | 0 | 0 | — |
case-13 | pass→pass | 16,457 | 2,051 | -88% | 1 | 1 | 0% | 2,520 | 535 | -79% | 0 | 0 | — |
case-14 | fail→pass | 5,907 | 2,405 | -59% | 1 | 1 | 0% | 890 | 508 | -43% | 0 | 0 | — |
case-15 | fail→pass | 10,525 | 2,395 | -77% | 1 | 1 | 0% | 1,588 | 592 | -63% | 0 | 0 | — |
case-16 | fail→pass | 7,883 | 2,214 | -72% | 1 | 1 | 0% | 1,301 | 458 | -65% | 0 | 0 | — |
case-17 | pass→pass | 8,097 | 2,517 | -69% | 1 | 1 | 0% | 1,056 | 620 | -41% | 0 | 0 | — |
case-18 | pass→pass | 8,051 | 2,815 | -65% | 1 | 1 | 0% | 1,209 | 669 | -45% | 0 | 0 | — |
case-19 | fail→pass | 13,054 | 1,696 | -87% | 1 | 1 | 0% | 1,988 | 471 | -76% | 0 | 0 | — |
case-20 | pass→pass | 9,674 | 7,514 | -22% | 1 | 1 | 0% | 1,678 | 1,224 | -27% | 0 | 0 | — |
case-21 | fail→fail | 4,512 | 3,372 | -25% | 1 | 1 | 0% | 559 | 621 | +11% | 0 | 0 | — |
case-22 | fail→pass | 3,639 | 6,238 | +71% | 1 | 1 | 0% | 556 | 1,068 | +92% | 0 | 0 | — |
case-23 | pass→fail | 5,498 | 6,538 | +19% | 1 | 1 | 0% | 989 | 1,268 | +28% | 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. 23 cases were attempted. The headline lift of +43 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is 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.