Apartment Finder

Find real listings, evaluate them, detect scams, vet neighborhoods, and flag lease traps.

When to Use

• Finding apartments in a specific city/area matching user criteria
• Evaluating a specific listing (especially Craigslist/FB Marketplace)
• Comparing units on true cost, not sticker rent
• Lease review before signing

When NOT to Use

• Home buying; active lease disputes (use legal-contract)

Search Methodology

When a user asks to find apartments, run a multi-site search and return specific units with addresses, prices, and direct links. Users do not want generic search page links — they want listings they can act on.

Search Process

1. Determine criteria from the user: city/neighborhood, bedrooms, max budget, preferences (waterfront, views, parking, pets, etc.)
2. Search all major sites in parallel using

   webSearch

   :
   • Zillow (

     zillow.com [city] [beds] bedroom [type] for rent [neighborhood] [preferences]

     )
   • Apartments.com (

     apartments.com [city] [beds] bedroom [neighborhood] rent

     )
   • Redfin (

     redfin.com [city] [beds] bedroom [type] for rent [neighborhood]

     )
   • Craigslist (

     craigslist [city] [beds] bedroom [type] [neighborhood] rent

     )
3. Fetch actual listing pages using

   webFetch

   on the most relevant result URLs to extract specific unit data (addresses, prices, sqft, features)
4. Present individual units — each listing needs: address, unit number, price, beds/baths, sqft, key features, and a direct link

Listing Type Preferences

Prefer individual units over managed-building "starting at" prices. Large apartment complexes advertise teaser "starting at" rates that are often for unavailable or least-desirable units. When possible:

• Search for condos for rent, townhouses for rent, and "by owner" listings
• Filter by property type = condo/townhouse on Redfin and Zillow
• On Craigslist, use the "housing by owner" filter
• On Apartments.com, look for "for rent by owner" listings
• Present the actual asking price for the specific unit, not "starting at" ranges
• When large buildings are included, note that the listed price is a floor and the user should expect higher for desirable units/floors.

Price Verification — Unit vs. Building

Listing sites often show a building's lowest "starting at" price that may be for a different unit type, floor plan, or even bedroom count than what the user asked for. Before presenting any listing:

1. Confirm the price is for the right bedroom count. A building page might say "$3,450+" but that's for a 1BR — the 2BR starts at $5,200. Always drill into the specific floor plan.
2. Confirm the unit actually exists at that price. "Starting at" prices on Apartments.com, Zillow building pages, and RentCafe are often for a single unit or a unit that's already leased. If you can see individual unit prices on the page, quote those instead.
3. Quote specific unit prices when available. If a building has Unit 4B at $5,250 and Unit 7A at $5,800, present those individually — don't say "$5,250+". The user can evaluate each one.
4. Flag uncertainty. If you can only find a building-wide range and not a specific unit price, say so explicitly: "Building lists 2BRs from $X,XXX but specific unit pricing requires contacting leasing office."
5. Watch for furnished vs. unfurnished. Some sites (Blueground, Zeus, corporate housing) list furnished units at 30-50% premiums. Always note if a price includes furnishing.

Output Format

Present listings in tiers based on value relative to the user's budget:

**[Address, Unit #]** — **$X,XXX/mo** | Xbd/Xba | X,XXX sq ft
- [Key feature 1], [key feature 2], [key feature 3]
- [Why this is good value / what stands out]
- [Direct link to listing](URL)

Group into:

• Best Value — significantly under budget, punches above its price
• Strong Mid-Range — solid quality at a fair price within budget
• Top of Budget — premium units near the ceiling that justify the spend

After the listings, include:

• A "Top Picks" summary (2-3 sentences on the best overall deals and why)
• Filtered search links for each site so the user can monitor new listings
• True cost notes (parking, utilities, fees) relevant to the market

Scam Warnings on Presented Listings

When presenting any listing, actively scan for red flags and warn the user inline. Do not present a suspicious listing without a warning attached.

Auto-flag if any of these are true:

• Price is 30%+ below neighborhood average for its size/type — add: "Price is well below market. Verify this is real before sending any money."
• Listing source is Craigslist or FB Marketplace with no verifiable landlord identity — add: "Listed on [platform] with no identity verification. Confirm ownership via county assessor before touring."
• Photos look professional but listing is "by owner" at a suspiciously low price — add: "Reverse-image-search the photos to confirm they aren't stolen from another listing."
• Listing says "no background check" or "no credit check" — add: "Legitimate landlords almost always run background/credit checks. Proceed with caution."
• Any mention of mailing keys, wiring money, or inability to show the unit in person — add: "DO NOT send money. This matches a known scam pattern."

For every Craigslist/FB listing presented, append this standard note:

Craigslist/FB listings have no identity verification. Before sending any deposit: tour the unit in person, verify the owner via county property records, and pay only by check to a verified business entity. Never pay via Zelle/Venmo/CashApp/wire/gift card.

For individual condo/owner listings from any site, note:

Private owner listing — verify ownership at [county] county assessor's website before signing anything or sending money.

Search Queries That Work

Tailor queries to each site's strengths:

[city] [neighborhood] [beds] bedroom condo for rent

[city] [neighborhood] [beds] bedroom [condo/apartment] for rent

[city] [neighborhood] [beds] bedrooms

[city metro] [beds]br [neighborhood] [preferences]

[city] [neighborhood] condos for rent [beds] bedrooms

Run at least 2 rounds of searches: a broad initial sweep, then targeted follow-ups on the most promising neighborhoods/buildings using

webFetch

Identifying "Underpriced" Listings

When the user wants deals or value:

• Search for average rent in the target neighborhood using

  webSearch

• Flag any listing priced 15%+ below neighborhood average for its size
• Look for: older buildings with recent renovations, move-in concessions (1 month free, weeks free), private owners pricing to fill quickly (vacant 30+ days), winter/off-season listings
• Check for concession-adjusted effective rent (e.g., 1 month free on 12-month lease = ~8% discount)

Scam Detection — Named Patterns

Craigslist and FB Marketplace have no identity verification — scammers exploit this. FB is worse now because aged accounts with fake friends buy false trust.

The "overseas landlord" script (most common): Owner is abroad for work/military/missionary trip → can't show unit but will "mail keys" after deposit → below-market rent → no background check needed → pushes Zelle/Venmo/CashApp → disappears. If any two of these appear together, stop.

The hijacked listing: Scammer copies photos/description from a real Zillow/Realtor.com listing, reposts at ~50% of real rent. Verification: reverse-image-search the photos (images.google.com); webSearch the address on Zillow — if it's listed for sale or at a much higher rent, the cheap listing is fake.

The "Zelle stopped working" escalation: After partial Zelle payment, claims a technical issue and pivots to gift cards. Gift cards = 100% scam, zero exceptions.

"[county name] property records"

FB profile tells: <50 friends OR hundreds of friends all from one foreign country; account created recently; no posts before the last few months.

Neighborhood Research (webSearch/webFetch)

crimemapping.com

spotcrime.com

[city] police department crime map

walkscore.com/[address]

howloud.com

greatschools.org

floodfactor.com

coveragemap.com

nsopw.gov

IRL check: Visit at 10pm on a Friday and 7am on a weekday. Noise, parking, and neighbor quality only show up in person.

Lease Red Flags

True Cost Comparison

Rent is 60-80% of actual cost. Build the full picture:

Always present true monthly cost alongside base rent when comparing listings.

Rent Negotiation

Works best in soft markets or with private landlords (corporate property managers have less flexibility).

• Leverage: longer lease term (18-24 mo), move-in during winter (Nov-Feb = lowest demand), proof of income >3× rent, 750+ credit, offer to start lease immediately on a vacant unit
• Ask for: $50-100/mo off, one month free (amortize it), waived app/admin fee, free parking spot, waived pet deposit
• Data: pull 3 comps from Zillow/Apartments.com at lower rent, bring printouts

Before Move-In

• Video walkthrough with timestamps — every wall, inside every cabinet, run every faucet. Email it to yourself AND the landlord same-day
• Test: all outlets, water pressure hot+cold, all appliances, HVAC in both modes, every window lock, smoke detectors

Interactive Map — Web App Visualization

After gathering listings, build a web app that displays all found apartments on an interactive map so the user can visually compare locations.

Requirements

• Color-coded markers by tier: green = Best Value, blue = Strong Mid-Range, orange = Top of Budget, red = flagged/suspicious
• Popup on each marker showing: address, price, beds/baths, sqft, key features, and a link to the listing
• Sidebar or panel with the full listing list — clicking a listing pans/zooms to its marker

Geocoding Addresses

Use the free Nominatim API (OpenStreetMap) to convert addresses to lat/lng — no API key required:

https://nominatim.openstreetmap.org/search?q={url_encoded_address}&format=json&limit=1

Rate limit: max 1 request/second. Batch geocode all listings before building the map, and hardcode the coordinates into the app.

Always generate the map alongside the text-based listing output — the map is a visual complement, not a replacement for the detailed listings.

Limitations

• Cannot access live MLS/listing feeds directly — use

  webSearch

  and

  webFetch

  to pull from public listing sites
• Listing data changes daily — always note the search date and tell users to verify availability
• Ownership verification requires user to check county records
• Lease law varies by state — flag issues, recommend local tenant-rights org for specifics