Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Compiles data from the knowledge base into a structured report formatted for the
.claude/skills/miosa-osa-report/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -45% | 0% |
> Generate structured reports from operation data and knowledge.
/report <topic> [--format <brief|detailed|executive>] [--period <week|month|quarter>]Compiles a structured report by searching the knowledge base, aggregating data points, and formatting for the target audience. The Lead agent owns report generation and ensures proper genre encoding.
--format.| Format | Audience | Length | Focus | |--------|----------|--------|-------| | brief | Executives, clients | 200 words | Key points + action | | detailed | Team leads, managers | 500-1000 words | Full analysis | | executive | C-suite, board | 100 words | Decisions needed |
bash# Weekly brief for client /report "project status" --format brief --period week # Detailed monthly analysis /report "performance metrics" --format detailed --period month # Executive summary /report "Q1 results" --format executive --period quarter
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-07 | fail→pass | 29,201 | 37,513 | +28% | 1 | 1 | 0% | 1,782 | 1,416 | -21% | 0 | 0 | — |
case-01 | fail→pass | 10,415 | 5,209 | -50% | 1 | 1 | 0% | 1,549 | 1,123 | -28% | 0 | 0 | — |
case-02 | fail→pass | 9,040 | 2,429 | -73% | 1 | 1 | 0% | 1,309 | 733 | -44% | 0 | 0 | — |
case-03 | fail→pass | 9,697 | 2,944 | -70% | 1 | 1 | 0% | 1,523 | 696 | -54% | 0 | 0 | — |
case-04 | fail→pass | 10,325 | 4,464 | -57% | 1 | 1 | 0% | 1,672 | 918 | -45% | 0 | 0 | — |
case-05 | fail→pass | 11,274 | 8,407 | -25% | 1 | 1 | 0% | 1,115 | 1,459 | +31% | 0 | 0 | — |
case-06 | fail→pass | 40,516 | 19,854 | -51% | 1 | 1 | 0% | 6,300 | 2,996 | -52% | 0 | 0 | — |
case-08 | fail→pass | 15,784 | 2,500 | -84% | 1 | 1 | 0% | 2,140 | 749 | -65% | 0 | 0 | — |
case-09 | fail→pass | 9,126 | 2,389 | -74% | 1 | 1 | 0% | 1,548 | 697 | -55% | 0 | 0 | — |
case-10 | pass→pass | 10,136 | 2,097 | -79% | 1 | 1 | 0% | 1,911 | 532 | -72% | 0 | 0 | — |
case-11 | fail→pass | 12,795 | 2,473 | -81% | 1 | 1 | 0% | 980 | 579 | -41% | 0 | 0 | — |
case-12 | pass→pass | 13,467 | 2,931 | -78% | 1 | 1 | 0% | 1,639 | 596 | -64% | 0 | 0 | — |
case-13 | pass→pass | 10,200 | 2,281 | -78% | 1 | 1 | 0% | 1,524 | 560 | -63% | 0 | 0 | — |
case-14 | fail→pass | 5,619 | 2,290 | -59% | 1 | 1 | 0% | 886 | 601 | -32% | 0 | 0 | — |
case-15 | fail→pass | 10,407 | 1,561 | -85% | 1 | 1 | 0% | 1,457 | 548 | -62% | 0 | 0 | — |
case-16 | fail→pass | 4,503 | 2,610 | -42% | 1 | 1 | 0% | 707 | 600 | -15% | 0 | 0 | — |
case-17 | fail→pass | 11,564 | 2,088 | -82% | 1 | 1 | 0% | 1,404 | 591 | -58% | 0 | 0 | — |
case-18 | pass→pass | 23,978 | 2,237 | -91% | 1 | 1 | 0% | 2,082 | 512 | -75% | 0 | 0 | — |
case-19 | fail→pass | 9,833 | 3,395 | -65% | 1 | 1 | 0% | 1,595 | 551 | -65% | 0 | 0 | — |
case-20 | fail→fail | 3,896 | 9,342 | +140% | 1 | 1 | 0% | 472 | 1,408 | +198% | 0 | 0 | — |
case-21 | pass→pass | 6,808 | 6,011 | -12% | 1 | 1 | 0% | 952 | 1,178 | +24% | 0 | 0 | — |
case-22 | pass→pass | 10,541 | 6,527 | -38% | 1 | 1 | 0% | 1,394 | 1,335 | -4% | 0 | 0 | — |
case-23 | pass→pass | 7,188 | 2,212 | -69% | 1 | 1 | 0% | 897 | 565 | -37% | 0 | 0 | — |
case-24 | pass→pass | 4,303 | 4,746 | +10% | 1 | 1 | 0% | 558 | 933 | +67% | 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. The headline lift of +63 percentage points is the difference between those two pass rates over the 24 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.