Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a firm, effective complaint letter that gets a resolution. Use when asked to write a complaint letter, complain to a company about a product/service, escalate poor service, or demand a refund/replacement. Produces a structured complaint — the facts, the impact, the specific resolution you want, and a deadline — in a firm, professional tone that's hard to ignore and easy to act on.
.claude/skills/mohitagw15856-complaint-letter/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 63% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 59% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 77% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 77% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 43% | 0% |
A complaint gets resolved when it's specific, reasonable, and makes the desired action obvious — not when it's angry. This skill writes a letter that lays out the facts, states exactly what you want, and gives a clear deadline, in a firm professional tone that a customer-service team can actually action.
Given "complain about a flight that was cancelled and they won't refund me", write the full letter anyway — infer the standard facts and a reasonable resolution, and bracket the specifics (dates, order/reference numbers, amounts) to fill in. Never hand back advice instead of the letter.
Ask for these only if they aren't already provided (else infer and bracket):
A ready-to-send letter:
Provide a short email version too, and notes on anything to confirm.
Consumer-advocacy correspondence practice — factual specificity, a concrete remedy, a reasonable deadline, and a stated escalation path.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 7,953 | 8,809 | +11% | 1 | 1 | 0% | 1,340 | 2,200 | +64% | 0 | 0 | — |
case-02 | pass→fail | 14,357 | 13,879 | -3% | 1 | 1 | 0% | 2,561 | 3,271 | +28% | 0 | 0 | — |
case-03 | pass→pass | 6,723 | 7,987 | +19% | 1 | 1 | 0% | 1,152 | 2,130 | +85% | 0 | 0 | — |
case-04 | fail→pass | 9,204 | 10,766 | +17% | 1 | 1 | 0% | 1,673 | 2,722 | +63% | 0 | 0 | — |
case-05 | pass→pass | 4,918 | 13,804 | +181% | 1 | 1 | 0% | 754 | 3,138 | +316% | 0 | 0 | — |
case-06 | pass→pass | 7,502 | 9,930 | +32% | 1 | 1 | 0% | 1,283 | 2,517 | +96% | 0 | 0 | — |
case-07 | fail→pass | 8,379 | 9,299 | +11% | 1 | 1 | 0% | 1,437 | 2,288 | +59% | 0 | 0 | — |
case-08 | pass→pass | 10,824 | 9,871 | -9% | 1 | 1 | 0% | 1,948 | 2,360 | +21% | 0 | 0 | — |
case-09 | fail→pass | 9,218 | 13,085 | +42% | 1 | 1 | 0% | 1,672 | 2,952 | +77% | 0 | 0 | — |
case-10 | pass→pass | 10,607 | 11,577 | +9% | 1 | 1 | 0% | 1,996 | 2,773 | +39% | 0 | 0 | — |
case-11 | fail→pass | 8,109 | 9,883 | +22% | 1 | 1 | 0% | 1,395 | 2,473 | +77% | 0 | 0 | — |
case-12 | pass→pass | 9,014 | 9,465 | +5% | 1 | 1 | 0% | 1,547 | 2,387 | +54% | 0 | 0 | — |
case-13 | fail→pass | 9,670 | 10,637 | +10% | 1 | 1 | 0% | 1,650 | 2,362 | +43% | 0 | 0 | — |
case-14 | fail→pass | 11,035 | 8,870 | -20% | 1 | 1 | 0% | 1,891 | 2,299 | +22% | 0 | 0 | — |
case-15 | fail→pass | 6,749 | 12,577 | +86% | 1 | 1 | 0% | 1,302 | 3,004 | +131% | 0 | 0 | — |
case-16 | pass→pass | 7,986 | 10,717 | +34% | 1 | 1 | 0% | 1,292 | 2,429 | +88% | 0 | 0 | — |
case-17 | fail→pass | 8,290 | 11,613 | +40% | 1 | 1 | 0% | 1,421 | 2,587 | +82% | 0 | 0 | — |
case-18 | pass→fail | 8,148 | 10,887 | +34% | 1 | 1 | 0% | 1,473 | 2,515 | +71% | 0 | 0 | — |
case-19 | pass→pass | 9,372 | 10,610 | +13% | 1 | 1 | 0% | 1,610 | 2,490 | +55% | 0 | 0 | — |
case-20 | pass→pass | 9,000 | 9,312 | +3% | 1 | 1 | 0% | 1,606 | 2,177 | +36% | 0 | 0 | — |
case-21 | fail→pass | 11,105 | 10,180 | -8% | 1 | 1 | 0% | 2,009 | 2,458 | +22% | 0 | 0 | — |
case-22 | fail→pass | 8,616 | 9,123 | +6% | 1 | 1 | 0% | 1,395 | 2,342 | +68% | 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 +36 percentage points is the difference between those two pass rates over the 22 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.