Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a compelling, accurate real-estate listing description. Use when asked to write a property listing, an MLS/Zillow description, a real-estate listing, or to make a property description more appealing. Produces a listing — a hook headline, a flowing description that sells the lifestyle and key features, a highlights list, and neighbourhood notes — accurate and Fair-Housing-compliant. Not legal advice.
.claude/skills/mohitagw15856-property-listing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 79% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 19% | 0% |
A great listing sells the feeling of living there while staying truthful — it leads with what makes the home special, paints the lifestyle, and gives buyers the facts they need to want a showing. This skill writes that description: appealing, scannable, and accurate, without the tired clichés or anything that crosses fair-housing lines.
> Note: this is a marketing aid, not legal advice. Listings are regulated — Fair Housing laws > prohibit language that indicates a preference or steers based on protected characteristics (race, religion, > familial status, disability, etc.), and claims must be truthful. Describe the property, not the ideal > buyer; have material claims and compliance reviewed per your jurisdiction/MLS rules.
Given the basics (beds/baths, key features), write the listing anyway — infer appealing, plausible framing from what's given, and mark any specific claim (confirm) (square footage, year, schools, HOA). Never invent facts (size, upgrades, permits) and never use buyer-preference language. Describe the home.
Ask for these only if they aren't already provided (else infer/flag to confirm):
Keep it truthful; mark figures to confirm.
Real-estate marketing practice — lifestyle-led, feature-accurate listings that are scannable and Fair-Housing-compliant.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 15,329 | 15,434 | +1% | 1 | 1 | 0% | 1,692 | 2,474 | +46% | 0 | 0 | — |
case-02 | fail→pass | 18,327 | 15,852 | -14% | 1 | 1 | 0% | 2,172 | 2,665 | +23% | 0 | 0 | — |
case-03 | fail→pass | 8,419 | 9,760 | +16% | 1 | 1 | 0% | 1,396 | 2,499 | +79% | 0 | 0 | — |
case-04 | pass→pass | 19,520 | 16,716 | -14% | 1 | 1 | 0% | 1,980 | 2,369 | +20% | 0 | 0 | — |
case-05 | pass→pass | 21,472 | 21,327 | -1% | 1 | 1 | 0% | 2,332 | 3,059 | +31% | 0 | 0 | — |
case-06 | pass→pass | 24,210 | 13,997 | -42% | 1 | 1 | 0% | 3,670 | 2,234 | -39% | 0 | 0 | — |
case-07 | fail→pass | 16,277 | 18,058 | +11% | 1 | 1 | 0% | 1,733 | 2,503 | +44% | 0 | 0 | — |
case-08 | fail→pass | 17,399 | 14,238 | -18% | 1 | 1 | 0% | 1,930 | 2,305 | +19% | 0 | 0 | — |
case-09 | fail→pass | 15,005 | 21,465 | +43% | 1 | 1 | 0% | 1,264 | 2,695 | +113% | 0 | 0 | — |
case-10 | fail→pass | 18,376 | 17,657 | -4% | 1 | 1 | 0% | 1,759 | 2,393 | +36% | 0 | 0 | — |
case-11 | fail→pass | 14,206 | 13,714 | -3% | 1 | 1 | 0% | 1,315 | 1,803 | +37% | 0 | 0 | — |
case-12 | fail→pass | 16,635 | 14,700 | -12% | 1 | 1 | 0% | 1,504 | 2,409 | +60% | 0 | 0 | — |
case-13 | fail→pass | 12,170 | 28,110 | +131% | 1 | 1 | 0% | 1,691 | 2,455 | +45% | 0 | 0 | — |
case-14 | fail→pass | 11,638 | 12,146 | +4% | 1 | 1 | 0% | 696 | 1,838 | +164% | 0 | 0 | — |
case-15 | fail→pass | 11,565 | 23,004 | +99% | 1 | 1 | 0% | 1,623 | 2,118 | +30% | 0 | 0 | — |
case-16 | fail→pass | 20,780 | 23,197 | +12% | 1 | 1 | 0% | 2,274 | 2,388 | +5% | 0 | 0 | — |
case-17 | fail→pass | 12,198 | 16,563 | +36% | 1 | 1 | 0% | 1,500 | 2,300 | +53% | 0 | 0 | — |
case-18 | fail→pass | 13,388 | 9,298 | -31% | 1 | 1 | 0% | 1,529 | 2,081 | +36% | 0 | 0 | — |
case-19 | fail→pass | 9,855 | 12,308 | +25% | 1 | 1 | 0% | 1,505 | 2,317 | +54% | 0 | 0 | — |
case-20 | fail→fail | 14,852 | 13,944 | -6% | 1 | 1 | 0% | 1,659 | 2,855 | +72% | 0 | 0 | — |
case-21 | fail→pass | 13,532 | 9,892 | -27% | 1 | 1 | 0% | 1,839 | 2,382 | +30% | 0 | 0 | — |
case-22 | fail→pass | 13,334 | 11,636 | -13% | 1 | 1 | 0% | 2,057 | 2,593 | +26% | 0 | 0 | — |
case-23 | fail→fail | 12,841 | 19,810 | +54% | 1 | 1 | 0% | 1,863 | 2,303 | +24% | 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 +78 percentage points is the difference between those two pass rates over the 23 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.