Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a resignation letter that closes a chapter without burning it — short, warm, legally clean, and silent on everything that doesn't belong in a permanent file. Use when asked write my resignation letter, how do I resign professionally, what do I say when I quit, or review my resignation email. Produces the letter itself, the tell-your-manager-first script, the timing plan, and the list of things that must NOT go in writing.
.claude/skills/mohitagw15856-resignation-letter/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | 63% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 71% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 30% | 0% |
A resignation letter has exactly one job: create a clean, dated record that you resigned, effective when. It is not the place for grievances, gratitude essays, or explanations — it goes in a permanent file, may be read by lawyers, and will outlive every feeling in it. This skill writes the three-sentence version that professionals use, plus the parts that actually matter and happen around the letter: the conversation before it, the timing, and the discipline about what stays verbal.
Ask for these if not provided:
Full text, ready to send — addressed, dated, 3–5 sentences]
Opening line · the two sentences on why (verbal-only version) · transition offer · what to say if pressed on destination or counteroffered]
Money-date check results → chosen date · tell order: manager → manager's plan for team/peers → letter · same-day-walkout contingency: what's backed up and returned before the conversation]
| Tempted to include | Why not | Where it belongs | |---|---|---|
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-18 | fail→pass | 20,670 | 22,106 | +7% | 1 | 1 | 0% | 2,025 | 3,300 | +63% | 0 | 0 | — |
case-19 | fail→pass | 16,034 | 17,830 | +11% | 1 | 1 | 0% | 1,727 | 2,956 | +71% | 0 | 0 | — |
case-20 | fail→pass | 19,155 | 20,565 | +7% | 1 | 1 | 0% | 1,898 | 2,923 | +54% | 0 | 0 | — |
case-11 | pass→pass | 16,380 | 13,704 | -16% | 1 | 1 | 0% | 1,742 | 2,988 | +72% | 0 | 0 | — |
case-12 | fail→fail | 20,857 | 23,460 | +12% | 1 | 1 | 0% | 2,111 | 3,421 | +62% | 0 | 0 | — |
case-13 | pass→pass | 20,766 | 17,910 | -14% | 1 | 1 | 0% | 2,264 | 3,438 | +52% | 0 | 0 | — |
case-01 | fail→pass | 17,677 | 25,387 | +44% | 1 | 1 | 0% | 2,785 | 3,763 | +35% | 0 | 0 | — |
case-02 | fail→pass | 19,410 | 17,676 | -9% | 1 | 1 | 0% | 2,835 | 3,676 | +30% | 0 | 0 | — |
case-03 | fail→fail | 21,449 | 25,348 | +18% | 1 | 1 | 0% | 3,230 | 3,691 | +14% | 0 | 0 | — |
case-04 | fail→fail | 12,661 | 14,143 | +12% | 1 | 1 | 0% | 1,661 | 3,205 | +93% | 0 | 0 | — |
case-05 | fail→pass | 19,253 | 16,601 | -14% | 1 | 1 | 0% | 2,339 | 3,555 | +52% | 0 | 0 | — |
case-06 | pass→pass | 22,663 | 20,298 | -10% | 1 | 1 | 0% | 2,687 | 3,369 | +25% | 0 | 0 | — |
case-07 | fail→pass | 15,151 | 17,141 | +13% | 1 | 1 | 0% | 2,182 | 3,489 | +60% | 0 | 0 | — |
case-08 | fail→fail | 23,102 | 22,151 | -4% | 1 | 1 | 0% | 2,995 | 3,518 | +17% | 0 | 0 | — |
case-09 | pass→pass | 22,246 | 17,000 | -24% | 1 | 1 | 0% | 2,401 | 3,383 | +41% | 0 | 0 | — |
case-10 | fail→pass | 19,211 | 16,351 | -15% | 1 | 1 | 0% | 2,071 | 3,684 | +78% | 0 | 0 | — |
case-14 | fail→fail | 15,045 | 15,256 | +1% | 1 | 1 | 0% | 2,252 | 3,381 | +50% | 0 | 0 | — |
case-15 | pass→fail | 21,466 | 33,702 | +57% | 1 | 1 | 0% | 2,296 | 3,385 | +47% | 0 | 0 | — |
case-16 | fail→fail | 20,208 | 21,176 | +5% | 1 | 1 | 0% | 2,035 | 3,333 | +64% | 0 | 0 | — |
case-17 | fail→pass | 18,957 | 20,917 | +10% | 1 | 1 | 0% | 2,237 | 3,089 | +38% | 0 | 0 | — |
case-21 | pass→fail | 10,688 | 13,302 | +24% | 1 | 1 | 0% | 942 | 2,317 | +146% | 0 | 0 | — |
case-22 | pass→pass | 13,426 | 10,635 | -21% | 1 | 1 | 0% | 1,325 | 2,465 | +86% | 0 | 0 | — |
case-23 | pass→pass | 12,907 | 14,172 | +10% | 1 | 1 | 0% | 1,953 | 2,852 | +46% | 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. The headline lift of +30 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 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.