Practical AI agent skills for real estate research, powered by RealtyAPI.
These skills teach your AI agent how to use the RealtyAPI REST API — and how to turn public property data into useful research: property briefs, listing shortlists, rental research, CMAs (comparative market analyses), investment deal screening, and agent shortlists.
One API key, 30+ property portals worldwide — Redfin, Realtor.com, Zillow, Trulia, Homes.com, Apartments.com (US), Rightmove, Zoopla (UK), Idealista, SeLoger, Funda (Europe), Property Finder (Middle East), Domain (Australia), Airbnb, Vrbo, and more.
npx skills add realtyapi/realtyapi-skillsWorks with Claude Code, Cursor, OpenAI Codex, GitHub Copilot, Gemini CLI, Windsurf, VS Code, and other agents that support the Agent Skills spec.
Or download any skills/<name>/SKILL.md and upload it to your AI tool manually.
Set your RealtyAPI key as an environment variable:
export REALTYAPI_API_KEY=rt_...Get a key at realtyapi.io. The full endpoint index is
public at realtyapi.io/llms.txt, and every
provider exposes a keyless OpenAPI spec at https://<provider>.realtyapi.io/openapi.json.
| Skill | Use it when you want to... | Output |
|---|---|---|
realtyapi-api |
Route a raw real estate data request to the right RealtyAPI endpoint | API calls, endpoint discovery, credit/pagination guidance |
property-research |
Get a complete profile of one property by address or listing URL | Property brief: details, history, value, comps, neighborhood |
listing-search |
Find for-sale or sold homes matching criteria in an area | Ranked listings table with source links |
rental-market-research |
Research long-term or short-term rentals and rent trends | Rental options + rent summary |
comparative-market-analysis |
Estimate what a property is worth from comps and market data | CMA report with a value range and reasoning |
investment-deal-finder |
Find and screen foreclosures, discounts, and rental-yield deals | Ranked deal candidates with equity/yield math |
agent-finder |
Find or evaluate real estate agents in an area or for a property | Vetted agent shortlist with evidence |
Here's a Redfin link — pull the full details, value estimate, and comps, and tell me if it's priced fairly.
Find 3-bed houses under $600k in Austin, sorted by price per square foot.
Run a CMA on 166 W 22nd St Unit 1D, New York, NY 10011 and give me a value range.
What does a 2-bed long-term rental cost in central Lisbon, and is rent rising?
Find foreclosure and price-reduced deals under $250k in Cuyahoga County and rank by equity.
Who are the most active listing agents in Scottsdale for homes around $1M?
realtyapi-api (data layer: auth, endpoint discovery via llms.txt + openapi.json)
│
▼
real estate workflows
├─ property-research
├─ listing-search
├─ rental-market-research
├─ comparative-market-analysis
├─ investment-deal-finder
└─ agent-finder
The workflow skills use realtyapi-api as the data layer for endpoint details. Each
workflow produces a useful artifact, not just raw JSON.
- Workflow-first, API-second — users ask for an outcome (a brief, a CMA), not an endpoint.
- Discover, don't guess — find endpoints in
llms.txt, then read the provider'sopenapi.jsonfor exact params before calling. Never hardcode/guess paths. - Public-data only — RealtyAPI returns public real estate data. Don't promise private/MLS-restricted fields.
- Match portal to market — route each country/region to the right portal; there's no single global search.
- Cited outputs — include the source listing URL for every property, agent, or comp.
- Credit-aware — most calls cost 1 request credit; warn before pricier endpoints and cap pagination loops.
- Estimates ≠ appraisals — present valuations as ranges with confidence, not guarantees.
MIT — see LICENSE.