Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when a task matches a bound-deck playbook trigger, or when the user corrects output produced from a playbook. Fetch playbooks; propose patches from corrections.
.claude/skills/hashgraph-online-agent-deck-playbooks/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✓→✗ | ▼ Worse | -67% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | -81% | 0% |
| case-22 | ✓→✓ | ＝ Same ✓ | -27% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | -66% | 0% |
| case-02 | ✗→✗ | ＝ Same ✗ | -62% | 0% |
triggers on get_bound_deck, then get_playbook before improvising. Playbook bodies live on the deck — do not mirror them into local skills.propose_playbook_patch with item ops (prefer one add_item to Gotchas/Checklist) and evidence.user_feedback_excerpt as a short verbatim quote.propose_playbook_patch with kind: "create" and a thin body (one gotcha is enough).update_playbook.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 14,235 | 9,204 | -35% | 1 | 1 | 0% | 1,339 | 451 | -66% | 0 | 0 | — |
case-02 | fail→fail | 16,621 | 10,376 | -38% | 1 | 1 | 0% | 1,786 | 676 | -62% | 0 | 0 | — |
case-03 | fail→fail | 9,399 | 11,346 | +21% | 1 | 1 | 0% | 600 | 738 | +23% | 0 | 0 | — |
case-04 | fail→fail | 13,233 | 12,508 | -5% | 1 | 1 | 0% | 1,197 | 448 | -63% | 0 | 0 | — |
case-05 | pass→pass | 41,333 | 11,582 | -72% | 1 | 1 | 0% | 5,939 | 1,139 | -81% | 0 | 0 | — |
case-06 | fail→fail | 23,866 | 20,640 | -14% | 1 | 1 | 0% | 999 | 2,114 | +112% | 0 | 0 | — |
case-07 | fail→fail | 18,794 | 13,970 | -26% | 1 | 1 | 0% | 2,099 | 553 | -74% | 0 | 0 | — |
case-08 | fail→fail | 8,975 | 9,717 | +8% | 1 | 1 | 0% | 1,246 | 564 | -55% | 0 | 0 | — |
case-09 | fail→fail | 22,664 | 16,611 | -27% | 1 | 1 | 0% | 2,767 | 1,306 | -53% | 0 | 0 | — |
case-10 | fail→fail | 20,704 | 13,460 | -35% | 1 | 1 | 0% | 3,595 | 636 | -82% | 0 | 0 | — |
case-11 | fail→fail | 9,575 | 9,837 | +3% | 1 | 1 | 0% | 497 | 528 | +6% | 0 | 0 | — |
case-12 | fail→fail | 5,394 | 9,802 | +82% | 1 | 1 | 0% | 903 | 639 | -29% | 0 | 0 | — |
case-13 | fail→fail | 15,401 | 9,213 | -40% | 1 | 1 | 0% | 1,448 | 418 | -71% | 0 | 0 | — |
case-14 | fail→fail | 13,306 | 8,710 | -35% | 1 | 1 | 0% | 1,491 | 525 | -65% | 0 | 0 | — |
case-15 | fail→fail | 11,219 | 15,141 | +35% | 1 | 1 | 0% | 1,936 | 614 | -68% | 0 | 0 | — |
case-16 | fail→fail | 11,978 | 7,118 | -41% | 1 | 1 | 0% | 1,640 | 1,052 | -36% | 0 | 0 | — |
case-17 | fail→fail | 17,119 | 9,402 | -45% | 1 | 1 | 0% | 1,328 | 455 | -66% | 0 | 0 | — |
case-18 | fail→fail | 15,407 | 12,258 | -20% | 1 | 1 | 0% | 1,696 | 513 | -70% | 0 | 0 | — |
case-19 | fail→fail | 37,206 | 22,710 | -39% | 1 | 1 | 0% | 1,147 | 899 | -22% | 0 | 0 | — |
case-20 | fail→fail | 9,455 | 10,271 | +9% | 1 | 1 | 0% | 666 | 471 | -29% | 0 | 0 | — |
case-21 | pass→fail | 11,477 | 9,265 | -19% | 1 | 1 | 0% | 1,346 | 438 | -67% | 0 | 0 | — |
case-22 | pass→pass | 16,779 | 12,129 | -28% | 1 | 1 | 0% | 2,230 | 1,619 | -27% | 0 | 0 | — |
case-23 | fail→fail | 2,746 | 7,602 | +177% | 1 | 1 | 0% | 228 | 406 | +78% | 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, and 3 counted toward the lift figure. The other 20 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. A headline lift is not published for this run.
Other measured skills in the registry, with their headline benchmark lift.