Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generates customized client onboarding checklists with phased tasks, ownership assignments, dependencies, acceptance criteria, and email templates. Adapts to consulting, SaaS, or agency engagement models.
.claude/skills/onewave-ai-onboarding-checklist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 34% | 0% |
| case-13 | ✓→✗ | ▼ Worse | 123% | 0% |
Generate a comprehensive, customized client onboarding checklist saved as onboarding-checklist.md in the current working directory. The checklist is actionable, phase-structured, and tailored to the specific engagement type.
references/output-template.md -- Full output structure: header block, four phases, handoff criteria, formatting rules.references/engagement-types.md -- How to classify Consulting vs SaaS vs Agency, and per-type adaptation guidelines.references/task-banks.md -- Reference task pools per engagement type and phase.references/email-templates.md -- The five required email templates, their structure, and content guidelines.references/scaling-and-pitfalls.md -- Timeline and team-size scaling rules, common pitfalls, and the final quality checklist.Gather these before generating. If any are missing, ask for them explicitly:
references/engagement-types.md. This classification drives the structure and tone of the entire checklist.references/task-banks.md, adapting language, deadlines, and owners to the specific engagement. Reflect tech-stack items in relevant tasks.references/scaling-and-pitfalls.md based on the specified duration.references/scaling-and-pitfalls.md based on team composition.references/email-templates.md.onboarding-checklist.md following references/output-template.md. Apply the per-type adaptations in references/engagement-types.md.references/scaling-and-pitfalls.md.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,429 | 28,208 | +72% | 1 | 1 | 0% | 3,123 | 6,064 | +94% | 0 | 0 | — |
case-02 | fail→fail | 21,426 | 39,378 | +84% | 1 | 1 | 0% | 3,378 | 7,515 | +122% | 0 | 0 | — |
case-03 | fail→fail | 30,967 | 7,901 | -74% | 1 | 1 | 0% | 5,494 | 1,911 | -65% | 0 | 0 | — |
case-04 | pass→pass | 15,762 | 10,871 | -31% | 1 | 1 | 0% | 2,731 | 2,577 | -6% | 0 | 0 | — |
case-05 | pass→pass | 13,396 | 9,548 | -29% | 1 | 1 | 0% | 2,454 | 2,330 | -5% | 0 | 0 | — |
case-06 | pass→pass | 14,318 | 17,537 | +22% | 1 | 1 | 0% | 2,549 | 3,929 | +54% | 0 | 0 | — |
case-07 | fail→pass | 11,341 | 5,645 | -50% | 1 | 1 | 0% | 2,200 | 1,593 | -28% | 0 | 0 | — |
case-08 | fail→pass | 14,415 | 4,638 | -68% | 1 | 1 | 0% | 2,530 | 1,436 | -43% | 0 | 0 | — |
case-09 | fail→fail | 13,613 | 31,015 | +128% | 1 | 1 | 0% | 2,378 | 6,754 | +184% | 0 | 0 | — |
case-10 | pass→fail | 27,880 | 57,909 | +108% | 1 | 1 | 0% | 5,224 | 7,002 | +34% | 0 | 0 | — |
case-11 | pass→pass | 26,165 | 43,604 | +67% | 1 | 1 | 0% | 4,389 | 6,777 | +54% | 0 | 0 | — |
case-12 | pass→pass | 17,746 | 38,577 | +117% | 1 | 1 | 0% | 3,392 | 6,785 | +100% | 0 | 0 | — |
case-13 | pass→fail | 17,327 | 92,101 | +432% | 1 | 1 | 0% | 3,038 | 6,787 | +123% | 0 | 0 | — |
case-14 | fail→fail | 16,544 | 110,162 | +566% | 1 | 1 | 0% | 2,978 | 6,768 | +127% | 0 | 0 | — |
case-15 | fail→pass | 20,382 | 5,928 | -71% | 1 | 1 | 0% | 3,713 | 1,679 | -55% | 0 | 0 | — |
case-16 | pass→fail | 13,629 | 29,177 | +114% | 1 | 1 | 0% | 2,481 | 6,780 | +173% | 0 | 0 | — |
case-17 | pass→fail | 34,777 | 137,711 | +296% | 1 | 1 | 0% | 6,231 | 6,983 | +12% | 0 | 0 | — |
case-18 | fail→fail | 16,205 | 16,844 | +4% | 1 | 1 | 0% | 3,053 | 3,678 | +20% | 0 | 0 | — |
case-19 | pass→fail | 9,962 | 69,951 | +602% | 1 | 1 | 0% | 1,897 | 2,287 | +21% | 0 | 0 | — |
case-20 | pass→pass | 14,513 | 229,662 | +1482% | 1 | 1 | 0% | 2,700 | 6,772 | +151% | 0 | 0 | — |
case-21 | pass→pass | 27,796 | 31,845 | +15% | 1 | 1 | 0% | 5,001 | 6,777 | +36% | 0 | 0 | — |
case-22 | pass→fail | 15,022 | 8,818 | -41% | 1 | 1 | 0% | 2,649 | 2,059 | -22% | 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 21 counted toward the lift figure. The other 1 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 -14 percentage points is the difference between those two pass rates over the 21 comparable cases. 6 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.