Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Challenge an over-assessed property tax bill — check whether your assessment is too high, build the evidence, and file the appeal before the deadline. Use when asked to appeal my property taxes, my property assessment is too high, lower my property tax, or is my home over-assessed. Produces an over-assessment check (comparables vs your valuation), the evidence pack to build, the appeal steps and the strict deadline to watch, a realistic savings estimate, and what to expect at a hearing — flaggin
.claude/skills/mohitagw15856-property-tax-appeal/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 33% | 0% |
Property tax is based on an assessed value that's often wrong — and a successful appeal can cut your bill for years. Most people never challenge it because they don't know how. This helps you check whether you're genuinely over-assessed, assemble the comparables and evidence that win appeals, and file before the (strict) deadline — while being clear the process is local and this isn't tax or legal advice.
Ask for these if not provided:
Over-assessed? vs comparables/condition → likely yes / borderline / probably not]. Evidence to build: comps · record errors · photos · valuation if needed]. File: with assessor/board], by deadline] — informal review vs formal appeal]. Worth: ~annual saving if reduced]. Hearing: lead with comps + your supported value; factual and brief.
> Process, deadlines, and grounds vary by jurisdiction. Not tax or legal advice — confirm with your local assessor's office, and consider a professional for high-value properties.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 42,510 | 26,726 | -37% | 1 | 1 | 0% | 3,637 | 3,845 | +6% | 0 | 0 | — |
case-02 | fail→pass | 35,059 | 28,030 | -20% | 1 | 1 | 0% | 4,242 | 4,030 | -5% | 0 | 0 | — |
case-03 | fail→pass | 26,140 | 27,371 | +5% | 1 | 1 | 0% | 3,658 | 4,508 | +23% | 0 | 0 | — |
case-04 | pass→pass | 18,243 | 27,086 | +48% | 1 | 1 | 0% | 2,871 | 3,711 | +29% | 0 | 0 | — |
case-16 | pass→pass | 24,777 | 16,870 | -32% | 1 | 1 | 0% | 2,306 | 2,743 | +19% | 0 | 0 | — |
case-05 | pass→pass | 22,591 | 23,391 | +4% | 1 | 1 | 0% | 2,361 | 3,147 | +33% | 0 | 0 | — |
case-06 | pass→pass | 21,618 | 21,363 | -1% | 1 | 1 | 0% | 2,479 | 3,284 | +32% | 0 | 0 | — |
case-07 | fail→pass | 61,067 | 18,958 | -69% | 1 | 1 | 0% | 2,986 | 3,147 | +5% | 0 | 0 | — |
case-08 | fail→pass | 24,976 | 19,972 | -20% | 1 | 1 | 0% | 2,657 | 3,532 | +33% | 0 | 0 | — |
case-09 | pass→pass | 18,508 | 18,534 | +0% | 1 | 1 | 0% | 2,100 | 2,934 | +40% | 0 | 0 | — |
case-10 | pass→pass | 18,901 | 16,749 | -11% | 1 | 1 | 0% | 1,760 | 2,726 | +55% | 0 | 0 | — |
case-11 | pass→pass | 23,811 | 20,995 | -12% | 1 | 1 | 0% | 2,556 | 3,178 | +24% | 0 | 0 | — |
case-12 | pass→pass | 17,485 | 17,945 | +3% | 1 | 1 | 0% | 1,895 | 2,788 | +47% | 0 | 0 | — |
case-13 | pass→pass | 19,604 | 25,672 | +31% | 1 | 1 | 0% | 2,584 | 3,888 | +50% | 0 | 0 | — |
case-14 | pass→pass | 22,653 | 23,591 | +4% | 1 | 1 | 0% | 2,552 | 3,777 | +48% | 0 | 0 | — |
case-15 | pass→pass | 19,641 | 18,465 | -6% | 1 | 1 | 0% | 2,476 | 3,642 | +47% | 0 | 0 | — |
case-17 | pass→pass | 15,242 | 14,680 | -4% | 1 | 1 | 0% | 2,586 | 3,630 | +40% | 0 | 0 | — |
case-18 | fail→fail | 13,485 | 11,776 | -13% | 1 | 1 | 0% | 1,885 | 2,806 | +49% | 0 | 0 | — |
case-19 | pass→pass | 16,753 | 13,579 | -19% | 1 | 1 | 0% | 1,951 | 2,313 | +19% | 0 | 0 | — |
case-20 | pass→pass | 16,300 | 14,462 | -11% | 1 | 1 | 0% | 2,242 | 3,316 | +48% | 0 | 0 | — |
case-21 | pass→pass | 13,364 | 22,399 | +68% | 1 | 1 | 0% | 2,839 | 4,225 | +49% | 0 | 0 | — |
case-22 | pass→pass | 17,938 | 17,894 | -0% | 1 | 1 | 0% | 2,837 | 3,560 | +25% | 0 | 0 | — |
case-23 | pass→pass | 20,031 | 18,976 | -5% | 1 | 1 | 0% | 2,786 | 3,438 | +23% | 0 | 0 | — |
case-24 | pass→pass | 16,423 | 15,033 | -8% | 1 | 1 | 0% | 2,175 | 3,288 | +51% | 0 | 0 | — |
case-25 | pass→pass | 5,950 | 9,227 | +55% | 1 | 1 | 0% | 1,126 | 2,460 | +118% | 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 +20 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.