Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Strategic advisory for proptech founders on real-estate segments, MLS/brokerage models, licensing, and business models. Use when scoping a proptech idea or when the user mentions proptech, MLS, brokerage, iBuyer, or RESPA.
.claude/skills/borghei-proptech-advisor/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 41% | 0% |
Strategic frameworks for property-technology founders, operators, and product leaders.
> Disclaimer: Frameworks only. Real estate is heavily regulated state-by-state — engage real-estate-licensed counsel for binding decisions.
proptech, real estate, real-estate, MLS, brokerage, broker, agent, iBuyer, property management, multifamily, commercial real estate, CRE, RESPA, fair housing, listings, transaction, escrow, title
bashpython scripts/market_segment_classifier.py description.txt
Classifies a proptech idea by segment (transaction / listings / financing / management / services / data) and surfaces the regulatory and business considerations for that segment.
references/proptech_segments.mdTime Estimate: 4-6 weeks for first scope.
references/mls_and_brokerage.mdTime Estimate: 4-12 weeks depending on path.
Time Estimate: 4-8 weeks.
Classifies a proptech business description into one or more proptech segments and surfaces the regulatory considerations for each.
bashpython scripts/market_segment_classifier.py description.txt python scripts/market_segment_classifier.py description.txt --json
Segments detected:
references/proptech_segments.md — Segments overview, regulatory exposure per segment, business model patternsreferences/mls_and_brokerage.md — How MLS works, IDX/VOW/RESO, broker licensing, partnership modelsassets/proptech_segment_assessment.md — Decision template for capturing segment, regulatory, GTM decisions| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 25,621 | 19,604 | -23% | 1 | 1 | 0% | 3,518 | 3,723 | +6% | 0 | 0 | — |
case-02 | pass→pass | 17,375 | 24,670 | +42% | 1 | 1 | 0% | 2,526 | 4,376 | +73% | 0 | 0 | — |
case-03 | pass→pass | 18,565 | 18,034 | -3% | 1 | 1 | 0% | 2,863 | 3,535 | +23% | 0 | 0 | — |
case-04 | pass→pass | 20,014 | 15,455 | -23% | 1 | 1 | 0% | 2,981 | 2,956 | -1% | 0 | 0 | — |
case-05 | pass→pass | 19,686 | 17,211 | -13% | 1 | 1 | 0% | 2,785 | 3,307 | +19% | 0 | 0 | — |
case-06 | fail→pass | 9,498 | 2,394 | -75% | 1 | 1 | 0% | 1,571 | 1,255 | -20% | 0 | 0 | — |
case-07 | fail→fail | 21,209 | 23,435 | +10% | 1 | 1 | 0% | 3,002 | 4,116 | +37% | 0 | 0 | — |
case-08 | fail→pass | 18,255 | 18,899 | +4% | 1 | 1 | 0% | 2,467 | 3,539 | +43% | 0 | 0 | — |
case-09 | fail→pass | 20,452 | 14,672 | -28% | 1 | 1 | 0% | 3,071 | 3,246 | +6% | 0 | 0 | — |
case-10 | pass→pass | 19,205 | 22,915 | +19% | 1 | 1 | 0% | 2,835 | 4,255 | +50% | 0 | 0 | — |
case-21 | fail→pass | 15,488 | 16,607 | +7% | 1 | 1 | 0% | 2,251 | 3,179 | +41% | 0 | 0 | — |
case-11 | fail→pass | 8,621 | 2,955 | -66% | 1 | 1 | 0% | 1,423 | 1,240 | -13% | 0 | 0 | — |
case-12 | fail→fail | 12,837 | 11,124 | -13% | 1 | 1 | 0% | 1,859 | 2,421 | +30% | 0 | 0 | — |
case-13 | fail→pass | 2,696 | 3,203 | +19% | 1 | 1 | 0% | 409 | 1,294 | +216% | 0 | 0 | — |
case-14 | pass→pass | 11,952 | 10,157 | -15% | 1 | 1 | 0% | 1,692 | 2,330 | +38% | 0 | 0 | — |
case-15 | fail→fail | 17,538 | 16,230 | -7% | 1 | 1 | 0% | 2,564 | 3,171 | +24% | 0 | 0 | — |
case-16 | fail→pass | 7,347 | 8,272 | +13% | 1 | 1 | 0% | 1,121 | 1,995 | +78% | 0 | 0 | — |
case-17 | fail→pass | 16,347 | 7,015 | -57% | 1 | 1 | 0% | 2,395 | 1,903 | -21% | 0 | 0 | — |
case-18 | fail→pass | 19,638 | 12,506 | -36% | 1 | 1 | 0% | 3,159 | 2,783 | -12% | 0 | 0 | — |
case-19 | pass→pass | 16,050 | 20,100 | +25% | 1 | 1 | 0% | 2,294 | 3,709 | +62% | 0 | 0 | — |
case-20 | pass→pass | 13,764 | 15,782 | +15% | 1 | 1 | 0% | 2,222 | 3,316 | +49% | 0 | 0 | — |
case-22 | fail→pass | 16,573 | 14,988 | -10% | 1 | 1 | 0% | 2,546 | 3,180 | +25% | 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 +50 percentage points is the difference between those two pass rates over the 22 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.