Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →The morning brief — one synthesis across all domains. Run at the start of the day or on demand.
.claude/skills/abaans-today/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 187% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -68% | 0% |
Produce today's brief across every active domain.
data/*.json (run refresh-domain first if stale)inputs/goals.md (if present)objects/ notes with open statuses (todo, doing, blocked, draft, in-progress)personal/outputs/brief-YYYY-MM-DD.md (typed doc, doc_kind: update)status: blocked, plus drafts still unsent.due in the next 14 days.done since the last brief.inputs/.brief instead of silently picking one.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 4,998 | 5,588 | +12% | 1 | 1 | 0% | 772 | 739 | -4% | 0 | 0 | — |
case-02 | fail→fail | 3,539 | 3,900 | +10% | 1 | 1 | 0% | 588 | 538 | -9% | 0 | 0 | — |
case-03 | fail→pass | 5,190 | 16,790 | +224% | 1 | 1 | 0% | 855 | 2,453 | +187% | 0 | 0 | — |
case-04 | fail→pass | 7,729 | 13,530 | +75% | 1 | 1 | 0% | 1,370 | 646 | -53% | 0 | 0 | — |
case-05 | pass→pass | 8,260 | 3,556 | -57% | 1 | 1 | 0% | 1,358 | 907 | -33% | 0 | 0 | — |
case-06 | fail→pass | 7,130 | 4,703 | -34% | 1 | 1 | 0% | 1,073 | 1,100 | +3% | 0 | 0 | — |
case-07 | fail→fail | 3,519 | 3,995 | +14% | 1 | 1 | 0% | 528 | 885 | +68% | 0 | 0 | — |
case-08 | fail→pass | 6,077 | 2,493 | -59% | 1 | 1 | 0% | 1,100 | 761 | -31% | 0 | 0 | — |
case-09 | pass→pass | 7,213 | 2,368 | -67% | 1 | 1 | 0% | 1,186 | 727 | -39% | 0 | 0 | — |
case-10 | pass→pass | 11,156 | 2,023 | -82% | 1 | 1 | 0% | 1,847 | 616 | -67% | 0 | 0 | — |
case-11 | fail→pass | 11,291 | 2,107 | -81% | 1 | 1 | 0% | 1,953 | 631 | -68% | 0 | 0 | — |
case-12 | fail→pass | 5,432 | 2,295 | -58% | 1 | 1 | 0% | 872 | 658 | -25% | 0 | 0 | — |
case-13 | fail→fail | 3,341 | 4,850 | +45% | 1 | 1 | 0% | 540 | 1,368 | +153% | 0 | 0 | — |
case-14 | fail→pass | 8,705 | 2,836 | -67% | 1 | 1 | 0% | 1,510 | 767 | -49% | 0 | 0 | — |
case-15 | fail→pass | 6,710 | 1,680 | -75% | 1 | 1 | 0% | 1,133 | 536 | -53% | 0 | 0 | — |
case-16 | pass→pass | 6,330 | 1,825 | -71% | 1 | 1 | 0% | 914 | 582 | -36% | 0 | 0 | — |
case-17 | pass→fail | 7,350 | 1,576 | -79% | 1 | 1 | 0% | 1,222 | 532 | -56% | 0 | 0 | — |
case-18 | fail→pass | 8,126 | 2,681 | -67% | 1 | 1 | 0% | 1,360 | 805 | -41% | 0 | 0 | — |
case-19 | pass→pass | 3,544 | 7,346 | +107% | 1 | 1 | 0% | 746 | 1,481 | +99% | 0 | 0 | — |
case-20 | pass→fail | 4,024 | 5,608 | +39% | 1 | 1 | 0% | 749 | 1,194 | +59% | 0 | 0 | — |
case-21 | fail→pass | 20,523 | 8,864 | -57% | 1 | 1 | 0% | 3,398 | 1,789 | -47% | 0 | 0 | — |
case-22 | pass→pass | 4,751 | 4,546 | -4% | 1 | 1 | 0% | 776 | 977 | +26% | 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 20 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 +36 percentage points is the difference between those two pass rates over the 20 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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/3/2026 | +22% |
Other measured skills in the registry, with their headline benchmark lift.