Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Design the buddy system that makes new-hire onboarding human — the buddy's actual job (context and safety, not training), the 30-day touchpoint plan, the ask-me-anything contract, and the buddy selection that avoids the two classic miscasts. Use when asked set up an onboarding buddy program, I'm buddying the new hire what do I do, our onboarding is docs with no humans, or the new person is drowning quietly. Produces the buddy role definition, the touchpoint schedule, the first-week script, and t
.claude/skills/mohitagw15856-onboarding-buddy-plan/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -9% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 38% | 0% |
Onboarding docs answer the documented questions; the buddy exists for the undocumented ones — "who actually decides this," "is it normal that nobody replied," "can I ask the VP directly or is that weird" — the questions new hires won't ask their manager because every one feels like an admission. The buddy's job is context and psychological safety, explicitly not training (that's the manager's and the docs' job — miscasting the buddy as trainer burns them out and confuses accountability). The plan makes the role concrete: selection, the touchpoint arc, the safety contract, and the line where buddy concerns route onward.
Ask for these if not provided:
Buddy owns: … · Manager owns: … · Docs own: … — stated to all three]
| When | Touchpoint | Its purpose | |---|---|---| Day 1 tour → week-1 dailies → weekly to 30 · the what-was-confusing question standing]
The unwritten norms list · the askable map · the agreements/handbook tour]
The day-one words · what routes to the manager (patterns-blocking-success, as advocacy) · what never does]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-12 | fail→pass | 14,959 | 13,581 | -9% | 1 | 1 | 0% | 2,557 | 3,451 | +35% | 0 | 0 | — |
case-19 | pass→pass | 14,125 | 13,245 | -6% | 1 | 1 | 0% | 2,168 | 3,170 | +46% | 0 | 0 | — |
case-11 | fail→pass | 17,292 | 17,995 | +4% | 1 | 1 | 0% | 3,086 | 4,337 | +41% | 0 | 0 | — |
case-01 | fail→fail | 20,591 | 20,056 | -3% | 1 | 1 | 0% | 3,563 | 4,736 | +33% | 0 | 0 | — |
case-02 | pass→pass | 29,434 | 20,179 | -31% | 1 | 1 | 0% | 4,445 | 4,577 | +3% | 0 | 0 | — |
case-03 | fail→pass | 24,855 | 16,002 | -36% | 1 | 1 | 0% | 4,269 | 3,898 | -9% | 0 | 0 | — |
case-04 | fail→pass | 16,185 | 13,068 | -19% | 1 | 1 | 0% | 2,597 | 3,316 | +28% | 0 | 0 | — |
case-05 | fail→pass | 15,121 | 14,517 | -4% | 1 | 1 | 0% | 2,743 | 3,797 | +38% | 0 | 0 | — |
case-06 | fail→pass | 13,756 | 14,712 | +7% | 1 | 1 | 0% | 2,498 | 3,596 | +44% | 0 | 0 | — |
case-07 | fail→fail | 14,649 | 14,542 | -1% | 1 | 1 | 0% | 2,599 | 3,768 | +45% | 0 | 0 | — |
case-08 | pass→pass | 16,254 | 16,357 | +1% | 1 | 1 | 0% | 2,556 | 3,817 | +49% | 0 | 0 | — |
case-09 | pass→pass | 12,972 | 14,274 | +10% | 1 | 1 | 0% | 2,108 | 3,351 | +59% | 0 | 0 | — |
case-10 | fail→pass | 11,058 | 15,206 | +38% | 1 | 1 | 0% | 1,846 | 3,490 | +89% | 0 | 0 | — |
case-13 | pass→pass | 14,979 | 15,605 | +4% | 1 | 1 | 0% | 2,618 | 3,776 | +44% | 0 | 0 | — |
case-14 | pass→pass | 20,743 | 20,587 | -1% | 1 | 1 | 0% | 3,053 | 4,364 | +43% | 0 | 0 | — |
case-15 | pass→pass | 16,661 | 14,587 | -12% | 1 | 1 | 0% | 2,496 | 3,369 | +35% | 0 | 0 | — |
case-16 | pass→pass | 15,475 | 18,001 | +16% | 1 | 1 | 0% | 2,612 | 3,929 | +50% | 0 | 0 | — |
case-17 | pass→fail | 13,499 | 8,363 | -38% | 1 | 1 | 0% | 2,022 | 2,443 | +21% | 0 | 0 | — |
case-18 | pass→pass | 14,933 | 13,826 | -7% | 1 | 1 | 0% | 2,376 | 3,395 | +43% | 0 | 0 | — |
case-20 | pass→pass | 13,759 | 12,484 | -9% | 1 | 1 | 0% | 2,299 | 3,224 | +40% | 0 | 0 | — |
case-21 | pass→pass | 13,693 | 12,710 | -7% | 1 | 1 | 0% | 2,144 | 3,122 | +46% | 0 | 0 | — |
case-22 | pass→pass | 14,636 | 11,245 | -23% | 1 | 1 | 0% | 2,236 | 2,983 | +33% | 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. The headline lift of +27 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.