Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Execute BioETL py-plan-bot profile for role-specific workflow and constraints.
.claude/skills/majiayu000-py-plan-bot/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-22 | ✓→✗ | ▼ Worse | -69% | 0% |
| case-20 | ✓→✓ | ＝ Same ✓ | 53% | 0% |
| case-21 | ✓→✓ | ＝ Same ✓ | 2% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | -7% | 0% |
| case-07 | ✗→✗ | ＝ Same ✗ | 17% | 0% |
Run the role-specific workflow as defined in the py-plan-bot profile.
../../agents/py-plan-bot.md../../agents/ORCHESTRATION.md../../../.ai/memory/agent-memory.md../../agents/py-plan-bot.md.../../agents/ORCHESTRATION.md.AGENTS.md and project constraints.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 30,779 | 70,645 | +130% | 1 | 1 | 0% | 4,592 | 4,248 | -7% | 0 | 0 | — |
case-07 | fail→fail | 24,339 | 18,545 | -24% | 1 | 1 | 0% | 2,887 | 3,375 | +17% | 0 | 0 | — |
case-02 | fail→fail | 30,027 | 30,371 | +1% | 1 | 1 | 0% | 3,934 | 4,400 | +12% | 0 | 0 | — |
case-03 | fail→fail | 20,940 | 22,141 | +6% | 1 | 1 | 0% | 3,168 | 4,299 | +36% | 0 | 0 | — |
case-04 | fail→fail | 29,612 | 27,195 | -8% | 1 | 1 | 0% | 4,620 | 4,229 | -8% | 0 | 0 | — |
case-05 | fail→fail | 22,421 | 32,508 | +45% | 1 | 1 | 0% | 3,318 | 4,637 | +40% | 0 | 0 | — |
case-06 | fail→fail | 34,008 | 25,453 | -25% | 1 | 1 | 0% | 5,406 | 3,962 | -27% | 0 | 0 | — |
case-08 | fail→fail | 33,659 | 34,273 | +2% | 1 | 1 | 0% | 4,523 | 5,248 | +16% | 0 | 0 | — |
case-09 | fail→fail | 23,916 | 27,816 | +16% | 1 | 1 | 0% | 3,245 | 4,272 | +32% | 0 | 0 | — |
case-10 | fail→fail | 22,205 | 35,226 | +59% | 1 | 1 | 0% | 3,250 | 3,798 | +17% | 0 | 0 | — |
case-11 | fail→fail | 22,330 | 28,455 | +27% | 1 | 1 | 0% | 3,019 | 4,224 | +40% | 0 | 0 | — |
case-12 | fail→fail | 33,406 | 30,833 | -8% | 1 | 1 | 0% | 4,742 | 4,954 | +4% | 0 | 0 | — |
case-13 | fail→fail | 19,080 | 17,353 | -9% | 1 | 1 | 0% | 3,147 | 736 | -77% | 0 | 0 | — |
case-14 | fail→fail | 27,015 | 24,124 | -11% | 1 | 1 | 0% | 4,313 | 4,470 | +4% | 0 | 0 | — |
case-15 | fail→fail | 26,071 | 20,527 | -21% | 1 | 1 | 0% | 3,415 | 3,812 | +12% | 0 | 0 | — |
case-16 | fail→fail | 22,940 | 22,103 | -4% | 1 | 1 | 0% | 3,120 | 4,083 | +31% | 0 | 0 | — |
case-17 | fail→fail | 26,043 | 24,037 | -8% | 1 | 1 | 0% | 3,289 | 4,455 | +35% | 0 | 0 | — |
case-18 | fail→fail | 24,682 | 31,729 | +29% | 1 | 1 | 0% | 3,283 | 4,998 | +52% | 0 | 0 | — |
case-19 | fail→fail | 28,873 | 30,470 | +6% | 1 | 1 | 0% | 3,900 | 4,358 | +12% | 0 | 0 | — |
case-20 | pass→pass | 27,305 | 34,799 | +27% | 1 | 1 | 0% | 5,222 | 7,991 | +53% | 0 | 0 | — |
case-21 | pass→pass | 13,160 | 12,895 | -2% | 1 | 1 | 0% | 1,257 | 1,283 | +2% | 0 | 0 | — |
case-22 | pass→fail | 6,965 | 10,199 | +46% | 1 | 1 | 0% | 982 | 307 | -69% | 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 -5 percentage points is the difference between those two pass rates over the 20 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.