Real Estate Analyzer

• Address, price, square footage, lot size
• Bedrooms, bathrooms, year built
• Property type (single-family, condo, townhouse, multi-family)
• Days on market, price history, price reductions

Condition Indicators:

• Age of major systems (roof, HVAC, water heater, electrical)
• Recent renovations or updates
• Foundation type and condition
• Photos analysis — look for staging tricks, unflattering angles, missing rooms

Red Flags:

• Significantly below market price (could indicate undisclosed issues)
• Frequent ownership changes (flipped too fast?)
• "As-is" or "investor special" language
• Missing disclosures or incomplete listing info
• High DOM (days on market) without price reduction

Step 2: Valuation Analysis

Pull comps — specific sources:

• webFetch Redfin sold filter: redfin.com/city/{id}/filter/include=sold-6mo — recently sold within 0.5mi, ±20% sqft, same beds
• Zillow Research (zillow.com/research/data/) — free CSV downloads of ZHVI (home value index) and ZORI (rent index) by ZIP, monthly back to 1996
• County assessor website — webSearch "{county name} property assessor {address}" for tax-assessed value, last sale price, permit history. Assessed value is typically 70-90% of market value.
• Adjust comps: ±$15-40/sqft for size delta, ±$5-15k per bedroom, ±10-20% for condition

Affordability math (compute, don't estimate):

# PITI at 30yr fixed — webSearch current rates (use Freddie Mac PMMS)
# PMMS reports percentage (e.g. 6.76), so divide by 100 first
P, r, n = loan_amount, annual_rate/100/12, 360
monthly_PI = P * (r*(1+r)**n) / ((1+r)**n - 1)

# + property tax (county rate × assessed value / 12)

# + homeowners insurance (~$150-250/mo, varies wildly by state)

# + PMI if <20% down (~0.5-1.0% of loan/yr)

• 28/36 rule: PITI <28% gross income, total debt <36%. Lenders stretch to 43% DTI — don't.
• Closing costs: 2-5% of purchase. Maintenance reserve: 1-2% of home value/yr.

Step 3: Neighborhood Analysis

webSearch/webFetch targets (name the source, don't be vague):

• Schools: greatschools.org/{state}/{city} — rating ≥7 protects resale value even if user has no kids
• Crime: crimemapping.com or spotcrime.com/{city} — check 6-month trend, not just snapshot. NeighborhoodScout for demographic overlay.
• Walk/Transit/Bike Score: walkscore.com/score/{address}
• Flood: msc.fema.gov/portal/search — Zone A/AE/V = mandatory flood insurance ($400-3,000+/yr, often kills deals)
• Market velocity: Redfin Data Center — median DOM, sale-to-list ratio, months of supply. <3 months supply = seller's market.
• Future development: webSearch "{city} planning commission agenda" + "{city} zoning map" — a highway expansion or apartment rezoning next door changes everything

Step 4: Investment Analysis — Run the Numbers

Quick-filter rules (kill deals fast):

• 1% rule: monthly rent ≥ 1% of purchase price. Dead in coastal/HCOL markets — there, 0.5-0.7% is realistic and you're betting on appreciation, not cash flow.
• 50% rule: operating expenses (NOT mortgage) eat ~50% of gross rent. Vacancy + repairs + management + taxes + insurance + capex reserve. Beginners always underestimate this.
• 70% rule (flips/BRRRR): max offer = (ARV × 0.70) − rehab cost. ARV = after-repair value from renovated comps.

Full underwriting (build in Python):

Gross rent (use Rentometer or Zillow ZORI for the ZIP)
− Vacancy (5-8% typical; 10% conservative)
− Property management (8-10% of collected rent)
− Repairs/maintenance (~8% of rent)
− CapEx reserve (~5% — roof/HVAC/water heater sinking fund)
− Taxes + insurance
= NOI (Net Operating Income)

Cap rate = NOI / purchase price
  → <4%: you're buying appreciation, not cash flow
  → 4-6%: typical for A/B-class in growth metros
  → 6-8%: solid cash flow, B/C-class
  → >10%: either a great deal or a war zone — verify crime data

NOI − annual debt service (P+I) = annual cash flow
Cash-on-cash = annual cash flow / total cash in (down pmt + closing + rehab)
  → Target 8%+ CoC. Below that, an index fund wins with zero tenants.

DSCR (what lenders check for investment loans):

• DSCR = NOI / annual debt service. Lenders want ≥1.20-1.25× (2025 standard). <1.0 means rent doesn't cover the mortgage.
• DSCR loans (2025): ~6.5-7.5% rate, qualify on property income not W-2, typical max 75% LTV. How investors scale past 10 conventional mortgages.

BRRRR stack: Buy distressed (hard money, 7-14 day close) → Rehab → Rent → Refinance at 75% of new ARV into DSCR loan → pull most capital out → Repeat. Only works if ARV × 0.75 ≥ purchase + rehab + holding costs.

Rent comps: webSearch Rentometer free tier, or pull Zillow rentals for the ZIP and compute median $/sqft for same bed count.

Step 5: Due Diligence Checklist

Before making an offer:

• Pre-approval letter from lender
• Professional home inspection ($300-500)
• Pest/termite inspection
• Title search for liens or encumbrances
• Survey (if boundaries unclear)
• Flood zone check (FEMA maps)
• Environmental concerns (radon, lead paint for pre-1978 homes)
• HOA review (financials, rules, pending assessments)
• Property tax history and assessment

Output Format

Always present key findings and recommendations as a plaintext summary in chat, even when also generating files. The user should be able to understand the results without opening any files.

# Property Analysis: [Address]

## Summary

- Asking Price: $XXX,XXX
- Estimated Fair Value: $XXX,XXX — [Over/Under/Fair priced by X%]
- Recommendation: [Strong Buy / Buy / Hold / Pass]

## Property Details
[Key facts table]

## Valuation
[Comps analysis, price per sqft comparison]

## Neighborhood
[Schools, safety, livability scores]

## Financial Analysis
[Monthly payment breakdown, investment returns if applicable]

## Risks & Concerns
[Red flags, upcoming expenses, market risks]

## Verdict
[2-3 sentence recommendation]

Best Practices

1. Asking price is marketing — only sold comps within 6 months matter
2. Model three scenarios — base case, 10% vacancy + 20% higher repairs, and "tenant trashes it year 1"
3. Permit history is free alpha — county assessor site shows pulled permits. No permits on an "updated kitchen" = unpermitted work = your liability.
4. Price/sqft is a blunt tool — lot size, corner lots, and basement finish skew it hard. Use for screening, not for offers.
5. Cap rate without appreciation — in a flat market, if cap rate < your mortgage rate, you're paying to own it

Interactive Map — Web App Visualization

After analyzing properties, build a web app that displays properties and relevant neighborhood data on an interactive map.

Property Markers

• Color-coded by recommendation: green = Strong Buy, blue = Buy, yellow = Hold, red = Pass
• Popup on each marker showing: address, asking price, estimated fair value, beds/baths, sqft, price/sqft, and recommendation
• Click to expand with key details: comp-adjusted value, cap rate (if investment), flood zone, school rating

Neighborhood Context Layers

Display relevant context around the properties:

• Sold comps — recent comparable sales as smaller markers, with sale price and date
• School locations with GreatSchools ratings (color-coded: green ≥7, yellow 4-6, red <4)
• Flood zones if any properties are in or near FEMA Zone A/AE/V
• Nearby amenities — transit, grocery, parks when walkability matters to the user

Geocoding

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