Property research
Real estate research skills for AI agents — property briefs, listing search, rentals, CMAs, investment deal screening, and agent finders, powered by RealtyAPI
npx -y skills add realtyapi/realtyapi-skills --skill property-researchAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use when the user wants a complete profile of a single property — by street address or a portal listing URL. Pulls full details, price/tax history, valuation/estimate, comparables, and neighborhood context (schools, climate, walkability) and turns them into a property brief. Triggers on "tell me about this house", "look up this address", "is this listing a good deal", "what's this property worth", "pull the details for this Zillow/Redfin/Rightmove link".
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
3.5 KB, as published. Nobody here has run it
Property Research
Overview
Produce a single, well-sourced brief on one property: what it is, its history, what it's likely worth, how it compares to nearby homes, and the neighborhood context — so the user can judge it quickly.
Use the realtyapi-api skill as the data layer: it
explains auth, endpoint discovery via llms.txt, and per-provider openapi.json.
When to Use
- "Tell me everything about 123 Main St."
- "Here's a Zillow/Redfin/Rightmove link — is it a good buy?"
- "What's this property worth and how does it compare to others nearby?"
- Pre-offer or pre-showing due diligence on a specific home.
Workflow
- Pick the portal from the address's country/market (see the catalog in
realtyapi-api). If the user pasted a listing URL, use that portal and its URL-based endpoint. - Resolve & fetch core details — call the property-details endpoint by address
or URL (
…byaddress/…byurl/detailsbyaddress). Useautocompletefirst if the endpoint needs an internal id. - History — pull price history and tax history (
pricehistory,taxinfo*,sold-prices) if available. - Valuation — pull the estimate/valuation (
zestimate*,estimates). - Comparables — pull comps/similar/nearby homes for context.
- Neighborhood (optional, if relevant) — schools, climate, walk/transit/bike.
- Synthesize into the brief below. Flag any field a portal didn't return rather than inventing it.
Output Format
# Property Brief: {address}
**Source:** {portal} — {listing URL} · Status: {for sale / sold / off-market}
## Snapshot
- Price / last list: {} · Beds/Baths: {} · Size: {} · Lot: {} · Year built: {} · Type: {}
## Estimated Value
- Estimate: {value or range} ({source})
- vs. list price: {over / under / in line}
## History
| Date | Event | Price |
|------|-------|------:|
## Comparables
| Address | Price | Beds/Baths | Size | $/sqft | Distance |
|---------|------:|-----------|-----:|-------:|---------:|
## Neighborhood (if pulled)
- Schools: {} · Walk/Transit: {} · Climate/risk: {}
## Assessment
- Priced {fairly/high/low} vs. estimate and comps because …
- Watch-outs: …
- Missing data: {fields no portal returned}
Common Pitfalls
- Don't present an estimate as an appraisal — it's a model output; show the range.
- Don't compare a US address against a non-US portal; match portal to market.
- Keep comps truly comparable (similar size/type/recency); note when they aren't.
- Warn before endpoints that cost more than 1 credit, and don't silently loop comps.