🏠 Real Estate Comp Puller — Apify MCP Server

An MCP (Model Context Protocol) server that pulls recently-sold comparable properties from Zillow and Redfin for any US address, deduplicates them, and returns a structured analysis with a suggested value range, avg $/sqft, and supporting comp list.

Built for homeowners and buyers who want a quick, AI-accessible answer to: "What's this property worth?"

---

What it does

Given an address (plus optional beds/baths/sqft), the get_comps tool:

1. Fetches recently-sold listings from Zillow and Redfin via Apify scrapers
2. Deduplicates results across sources
3. Filters by bed/bath match (±1) and sqft tolerance (default ±20%)
4. Returns:
   • Suggested value range (±5% of avg comp price)
   • Avg & median sale price
   • Avg price per sqft
   • Avg days on market
   • Full comp list sorted by sold date

---

Project structure

real-estate-comp-puller/
├── .actor/
│   ├── actor.json          # Apify Actor config (standby mode, MCP path)
│   ├── Dockerfile          # Container build
│   └── pay_per_event.json  # PPE pricing ($0.25/report)
├── src/
│   ├── __init__.py
│   ├── __main__.py         # Entrypoint
│   ├── main.py             # MCP server + Starlette app
│   ├── zillow.py           # Zillow scraper via Apify
│   ├── redfin.py           # Redfin scraper via Apify
│   └── analysis.py         # Dedup, filter, and analysis logic
├── requirements.txt
└── README.md

---

Local development

1. Install deps

$pip install -r requirements.txt

2. Set env vars

export APIFY_TOKEN=your_apify_token_here
export APIFY_META_ORIGIN=STANDBY
export ACTOR_STANDBY_PORT=3000

3. Run

$python -m src

The MCP server will be available at: http://localhost:3000/mcp

---

Deploy to Apify

1. Install Apify CLI

npm install -g apify-cli
apify login

2. Push the Actor

$apify push

3. Enable Standby mode

In Apify Console → your Actor → Settings → enable Standby mode.

4. Set your APIFY_TOKEN env var

In Apify Console → your Actor → Settings → Environment variables:

• APIFY_TOKEN = your token (needed to call Zillow/Redfin sub-actors)

5. Connect from Claude or any MCP client

{
  "mcpServers": {
    "real-estate-comp-puller": {
      "type": "http",
      "url": "https://YOUR_USERNAME--real-estate-comp-puller.apify.actor/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_APIFY_TOKEN"
      }
    }
  }
}

---

Monetization

This Actor uses Pay Per Event (PPE) pricing at $0.25 per comp report.

• Users are charged once per successful get_comps call that returns results
• You earn 80% → ~$0.20/report after Apify's 20% commission
• Platform costs (Zillow + Redfin actor runs, proxies) ~$0.03–0.07/run
• Net ~$0.13–0.17 per report

To activate monetization:

1. Go to Apify Console → your Actor → Publication
2. Select Pay Per Event
3. The events are pre-configured in .actor/pay_per_event.json

---

Example prompt

"Pull comps for 456 Oak Ave, Austin TX 78704 — 3 bed, 2 bath, 1800 sqft"

Example response:

{
  "comp_count": 7,
  "summary": {
    "suggested_value_range": { "low": "$418,500", "high": "$461,900" },
    "avg_sale_price": "$440,200",
    "median_sale_price": "$435,000",
    "avg_price_per_sqft": "$241.50",
    "avg_days_on_market": 18.3,
    "sources": ["zillow", "redfin"]
  },
  "comps": [...]
}

---

Roadmap

• Geocoding for true radius-based filtering (Google Maps / Nominatim)
• Realtor.com as a third source
• Price history trend chart
• Neighborhood median price trend (90-day, 180-day)
• Free tier: 1 free report, then PPE kicks in