Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Sibyl 乐观分析者 agent - 从积极角度分析实验结果
.claude/skills/majiayu000-sibyl-optimist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 392% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -82% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-07 | ✓→✗ | ▼ Worse | -60% | 0% |
| case-14 | ✓→✗ | ▼ Worse | -85% | 0% |
!SIBYL_WORKSPACE="$ARGUMENTS[0]" .venv/bin/python3 -c "from sibyl.orchestrate import render_skill_prompt; import os; ws = os.environ.get('SIBYL_WORKSPACE', ''); print(render_skill_prompt('optimist', workspace_path=ws))"
AGENT_NAME: sibyl-optimist AGENT_TIER: sibyl-light SIBYL_ROOT: /Users/cwan0785/sibyl-system
Workspace path: $ARGUMENTS0]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 14,281 | 21,195 | +48% | 1 | 1 | 0% | 1,392 | 3,440 | +147% | 0 | 0 | — |
case-02 | fail→fail | 15,263 | 19,134 | +25% | 1 | 1 | 0% | 246 | 918 | +273% | 0 | 0 | — |
case-03 | fail→fail | 16,195 | 16,924 | +5% | 1 | 1 | 0% | 409 | 433 | +6% | 0 | 0 | — |
case-04 | fail→pass | 11,835 | 32,114 | +171% | 1 | 1 | 0% | 1,089 | 5,362 | +392% | 0 | 0 | — |
case-05 | pass→pass | 10,827 | 8,528 | -21% | 1 | 1 | 0% | 995 | 517 | -48% | 0 | 0 | — |
case-06 | fail→pass | 21,876 | 7,742 | -65% | 1 | 1 | 0% | 2,783 | 494 | -82% | 0 | 0 | — |
case-07 | pass→fail | 18,452 | 18,314 | -1% | 1 | 1 | 0% | 1,995 | 789 | -60% | 0 | 0 | — |
case-08 | pass→pass | 24,024 | 38,538 | +60% | 1 | 1 | 0% | 2,938 | 4,046 | +38% | 0 | 0 | — |
case-14 | pass→fail | 20,475 | 11,503 | -44% | 1 | 1 | 0% | 2,957 | 436 | -85% | 0 | 0 | — |
case-09 | fail→pass | 24,408 | 35,327 | +45% | 1 | 1 | 0% | 3,155 | 4,795 | +52% | 0 | 0 | — |
case-10 | pass→fail | 20,514 | 12,180 | -41% | 1 | 1 | 0% | 2,442 | 405 | -83% | 0 | 0 | — |
case-11 | fail→fail | 10,468 | 18,881 | +80% | 1 | 1 | 0% | 272 | 327 | +20% | 0 | 0 | — |
case-12 | pass→fail | 21,718 | 5,333 | -75% | 1 | 1 | 0% | 2,705 | 301 | -89% | 0 | 0 | — |
case-13 | fail→fail | 21,309 | 12,079 | -43% | 1 | 1 | 0% | 2,507 | 496 | -80% | 0 | 0 | — |
case-15 | pass→fail | 22,163 | 12,021 | -46% | 1 | 1 | 0% | 2,935 | 289 | -90% | 0 | 0 | — |
case-16 | pass→fail | 25,797 | 13,439 | -48% | 1 | 1 | 0% | 3,257 | 438 | -87% | 0 | 0 | — |
case-17 | fail→fail | 18,227 | 8,023 | -56% | 1 | 1 | 0% | 2,950 | 440 | -85% | 0 | 0 | — |
case-18 | fail→fail | 21,990 | 17,764 | -19% | 1 | 1 | 0% | 2,695 | 630 | -77% | 0 | 0 | — |
case-19 | pass→fail | 21,978 | 18,219 | -17% | 1 | 1 | 0% | 3,698 | 672 | -82% | 0 | 0 | — |
case-20 | fail→fail | 12,171 | 15,603 | +28% | 1 | 1 | 0% | 723 | 960 | +33% | 0 | 0 | — |
case-21 | pass→pass | 36,761 | 58,636 | +60% | 1 | 1 | 0% | 5,065 | 8,342 | +65% | 0 | 0 | — |
case-22 | fail→fail | 25,763 | 41,503 | +61% | 1 | 1 | 0% | 4,056 | 6,792 | +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. 22 cases were attempted, and 8 counted toward the lift figure. The other 14 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 -18 percentage points is the difference between those two pass rates over the 8 comparable cases. 7 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.
Other measured skills in the registry, with their headline benchmark lift.