Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a letter to dispute an incorrect charge, bill, or record. Use when asked to dispute a credit-card charge, contest a bill or invoice, challenge a credit-report error, or formally dispute a fee. Produces a clear dispute letter — what's being disputed, why it's wrong, the evidence, and the correction requested — in the firm, paper-trail tone these situations need.
.claude/skills/mohitagw15856-dispute-letter/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 53% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 60% | 0% |
Disputes are won on a clear paper trail: state precisely what's wrong, attach the evidence, and request a specific correction in writing. This skill writes that letter so it's easy for the other side to verify and fix — and so you have a dated record if it escalates.
> Note: this is a drafting aid, not legal or financial advice. Deadlines and rights vary by > jurisdiction and provider (e.g. billing-error and credit-reporting rules); confirm the process and time > limits with the provider or a qualified advisor, and keep copies of everything.
Given "dispute a $90 charge I didn't authorise", write the full letter anyway — structure the dispute and bracket the specifics (account/reference numbers, dates, amounts) to fill in. Note where supporting evidence should be attached. Never withhold the letter for missing detail.
Ask for these only if they aren't already provided (else bracket to fill in):
Provide a short version for an online dispute form, and notes on documents to attach and any deadline to confirm.
Consumer dispute practice — precise identification, evidence-backed reasoning, a specific requested correction, and a documented paper trail.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | fail→pass | 10,875 | 10,508 | -3% | 1 | 1 | 0% | 1,750 | 2,304 | +32% | 0 | 0 | — |
case-14 | pass→pass | 9,693 | 12,016 | +24% | 1 | 1 | 0% | 1,622 | 2,799 | +73% | 0 | 0 | — |
case-01 | fail→fail | 10,707 | 12,723 | +19% | 1 | 1 | 0% | 1,977 | 2,979 | +51% | 0 | 0 | — |
case-02 | fail→pass | 13,808 | 12,007 | -13% | 1 | 1 | 0% | 2,245 | 2,952 | +31% | 0 | 0 | — |
case-03 | fail→fail | 14,692 | 12,760 | -13% | 1 | 1 | 0% | 2,602 | 3,010 | +16% | 0 | 0 | — |
case-04 | fail→pass | 9,879 | 9,334 | -6% | 1 | 1 | 0% | 1,631 | 2,495 | +53% | 0 | 0 | — |
case-05 | pass→fail | 8,615 | 10,105 | +17% | 1 | 1 | 0% | 1,607 | 2,782 | +73% | 0 | 0 | — |
case-06 | pass→pass | 7,700 | 8,992 | +17% | 1 | 1 | 0% | 1,516 | 2,472 | +63% | 0 | 0 | — |
case-07 | pass→pass | 7,150 | 10,690 | +50% | 1 | 1 | 0% | 1,326 | 2,784 | +110% | 0 | 0 | — |
case-15 | fail→pass | 11,459 | 14,445 | +26% | 1 | 1 | 0% | 1,926 | 3,455 | +79% | 0 | 0 | — |
case-08 | pass→pass | 9,817 | 10,199 | +4% | 1 | 1 | 0% | 1,813 | 2,618 | +44% | 0 | 0 | — |
case-09 | fail→pass | 10,900 | 12,662 | +16% | 1 | 1 | 0% | 2,040 | 3,255 | +60% | 0 | 0 | — |
case-10 | pass→pass | 8,756 | 11,902 | +36% | 1 | 1 | 0% | 1,524 | 2,991 | +96% | 0 | 0 | — |
case-11 | fail→fail | 11,914 | 12,479 | +5% | 1 | 1 | 0% | 2,035 | 2,786 | +37% | 0 | 0 | — |
case-12 | pass→pass | 9,180 | 10,200 | +11% | 1 | 1 | 0% | 1,659 | 2,644 | +59% | 0 | 0 | — |
case-16 | fail→pass | 7,370 | 10,521 | +43% | 1 | 1 | 0% | 1,312 | 2,832 | +116% | 0 | 0 | — |
case-17 | fail→pass | 10,739 | 10,752 | +0% | 1 | 1 | 0% | 1,812 | 2,525 | +39% | 0 | 0 | — |
case-18 | pass→pass | 8,235 | 10,751 | +31% | 1 | 1 | 0% | 1,402 | 2,414 | +72% | 0 | 0 | — |
case-19 | fail→fail | 12,439 | 17,017 | +37% | 1 | 1 | 0% | 1,994 | 3,721 | +87% | 0 | 0 | — |
case-20 | fail→fail | 14,234 | 15,199 | +7% | 1 | 1 | 0% | 2,188 | 3,181 | +45% | 0 | 0 | — |
case-21 | fail→fail | 11,578 | 14,590 | +26% | 1 | 1 | 0% | 1,936 | 3,200 | +65% | 0 | 0 | — |
case-22 | fail→fail | 24,585 | 19,664 | -20% | 1 | 1 | 0% | 3,604 | 3,807 | +6% | 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.