Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Get a retailer to match a lower price you found — or refund the difference when the price drops right after you bought — with the request written and the exact proof to show. Use when asked for a price match, I found it cheaper somewhere else, the price dropped after I bought it, or can I get a price adjustment. Produces the policy-aware request, the evidence that qualifies, the eligibility read (what usually counts vs excludes), and a fallback if they won't match.
.claude/skills/mohitagw15856-price-match-request/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 8% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 107% | 0% |
| case-25 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 11% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 16% | 0% |
Two easy wins most people leave on the table: matching a lower price before you buy, and claiming the difference when a price drops days after you bought. Both usually hinge on a written policy with a short window and specific rules. This works out whether your case likely qualifies, writes the ask, and tells you exactly what to screenshot — so a two-minute message can save real money.
Ask for these if not provided:
Likely eligible? yes / borderline / probably not] — because same-item + in-window + authorized-seller check].
The request > Item + exact price seen + where + date + the specific ask: match / refund the difference], per your price-match / price-adjustment] policy. Proof attached.
Screenshot / attach: the lower price · the item page (model/SKU visible) · in-stock status · the date · your receipt or order #
If they say no
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 20,931 | 15,131 | -28% | 1 | 1 | 0% | 2,252 | 2,492 | +11% | 0 | 0 | — |
case-02 | fail→pass | 20,747 | 14,357 | -31% | 1 | 1 | 0% | 2,629 | 2,845 | +8% | 0 | 0 | — |
case-03 | pass→pass | 23,022 | 16,480 | -28% | 1 | 1 | 0% | 2,697 | 3,117 | +16% | 0 | 0 | — |
case-04 | pass→pass | 13,912 | 16,420 | +18% | 1 | 1 | 0% | 1,451 | 2,663 | +84% | 0 | 0 | — |
case-05 | pass→pass | 17,146 | 17,786 | +4% | 1 | 1 | 0% | 1,810 | 2,727 | +51% | 0 | 0 | — |
case-06 | pass→pass | 15,442 | 13,714 | -11% | 1 | 1 | 0% | 1,328 | 2,442 | +84% | 0 | 0 | — |
case-07 | pass→pass | 12,674 | 20,150 | +59% | 1 | 1 | 0% | 1,644 | 2,463 | +50% | 0 | 0 | — |
case-08 | pass→pass | 19,793 | 10,437 | -47% | 1 | 1 | 0% | 1,893 | 2,844 | +50% | 0 | 0 | — |
case-09 | pass→pass | 21,169 | 17,441 | -18% | 1 | 1 | 0% | 2,245 | 2,901 | +29% | 0 | 0 | — |
case-10 | pass→pass | 14,588 | 15,270 | +5% | 1 | 1 | 0% | 1,438 | 2,465 | +71% | 0 | 0 | — |
case-11 | pass→pass | 16,046 | 15,481 | -4% | 1 | 1 | 0% | 1,367 | 2,296 | +68% | 0 | 0 | — |
case-12 | fail→fail | 14,550 | 23,371 | +61% | 1 | 1 | 0% | 1,259 | 2,491 | +98% | 0 | 0 | — |
case-13 | pass→pass | 20,469 | 16,528 | -19% | 1 | 1 | 0% | 2,012 | 2,744 | +36% | 0 | 0 | — |
case-14 | pass→pass | 14,875 | 19,263 | +29% | 1 | 1 | 0% | 1,504 | 2,890 | +92% | 0 | 0 | — |
case-15 | fail→pass | 10,422 | 15,601 | +50% | 1 | 1 | 0% | 1,075 | 2,223 | +107% | 0 | 0 | — |
case-16 | pass→pass | 15,190 | 15,317 | +1% | 1 | 1 | 0% | 1,567 | 2,539 | +62% | 0 | 0 | — |
case-17 | fail→fail | 18,695 | 18,346 | -2% | 1 | 1 | 0% | 2,136 | 2,879 | +35% | 0 | 0 | — |
case-24 | pass→pass | 7,818 | 14,337 | +83% | 1 | 1 | 0% | 1,308 | 2,341 | +79% | 0 | 0 | — |
case-18 | pass→pass | 15,336 | 16,002 | +4% | 1 | 1 | 0% | 1,548 | 2,605 | +68% | 0 | 0 | — |
case-19 | pass→pass | 18,207 | 13,909 | -24% | 1 | 1 | 0% | 1,171 | 2,138 | +83% | 0 | 0 | — |
case-20 | pass→pass | 16,346 | 17,061 | +4% | 1 | 1 | 0% | 1,583 | 2,736 | +73% | 0 | 0 | — |
case-21 | pass→pass | 21,717 | 15,530 | -28% | 1 | 1 | 0% | 2,263 | 2,726 | +20% | 0 | 0 | — |
case-22 | pass→pass | 16,443 | 16,548 | +1% | 1 | 1 | 0% | 1,572 | 2,575 | +64% | 0 | 0 | — |
case-23 | pass→pass | 11,860 | 33,756 | +185% | 1 | 1 | 0% | 1,800 | 2,749 | +53% | 0 | 0 | — |
case-25 | fail→pass | 20,246 | 15,791 | -22% | 1 | 1 | 0% | 2,458 | 3,220 | +31% | 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. 25 cases were attempted. The headline lift of +12 percentage points is the difference between those two pass rates over the 25 comparable cases.
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.