Real Estate AI Agent System

AI-Powered Solution for Real Estate Public Data Extraction

🌟 Overview

Real Estate AI Agent System is a Python-based solution that leverages AI agents and Bright Data's Model Context Protocol (MCP) server to extract, process, and deliver structured real estate property data from multiple sources.

• Automates public property data extraction from real estate websites like Zillow, Realtor.com, Redfin, and more
• Integrates with Bright Data proxies for robust anti-bot and geo-unblocking
• Uses Nebius Qwen LLM for adaptive, schema-validated property data extraction
• Outputs results as structured JSON for analytics or downstream applications

Table of Contents

• ✨ Features
• 🚀 Quickstart
• 🔧 Environment Setup
• 💡 Usage Example
• 📈 Key Capabilities
• 🔒 Security Best Practices

✨ Features

• Intelligent AI Agents: Uses CrewAI and LLM for adaptive data extraction and property detail parsing.
• Bright Data Integration: Seamless support for proxy rotation, CAPTCHA solving via MCP server.
• Strict JSON Schema: Always returns result in snake_case, schema-validated JSON.
• Plug-and-Play: Spin up an advanced real estate data pipeline in minutes.
• Cross-Platform: Python 3.9; requires Node.js for Bright Data MCP server.

🚀 Quickstart

1. Clone this repository

   git clone https://github.com/brightdata-com/real-estate-ai-agents.git
   cd real-estate-ai-agents
   

🔧 Environment Setup

Prerequisites

• Python 3.9+
• Node.js + npm (for Bright Data MCP server)
• Bright Data account with API token
• Nebius AI API key

Virtual Environment

macOS/Linux

python3.9 -m venv venv
source venv/bin/activate

Windows

python3.9 -m venv venv
.\venv\Scripts\activate

Install Dependencies

pip install "crewai-tools[mcp]" crewai mcp python-dotenv pandas

Add Environment Variables

Create a .env file in your project directory with the following:

BRIGHT_DATA_API_TOKEN="your_api_token_here"
WEB_UNLOCKER_ZONE="your_web_unlocker_zone"
BROWSER_ZONE="your_browser_zone"
NEBIUS_API_KEY="your_nebius_api_key"

💡 Usage Example

To run the agent:

python real_estate_agents.py

If successful, the script will extract property data from a real estate listing and output result like:

{
  "address": "123 Main Street, City, State 12345",
  "price": "$450,000",
  "bedrooms": 3,
  "bathrooms": 2,
  "square_feet": 1850,
  "lot_size": "0.25 acres",
  "year_built": 1995,
  "property_type": "Single Family Home",
  "listing_agent": "John Doe, ABC Realty",
  "days_on_market": 45,
  "mls_number": "MLS123456",
  "description": "Beautiful home with updated kitchen...",
  "image_urls": ["https://example.com/image1.jpg", "https://example.com/image2.jpg"],
  "neighborhood": "Downtown Historic District"
}

📈 Key Capabilities

• Extracts address, price, bedrooms, bathrooms, square footage, lot size, year built, property type, listing agent, days on market, MLS number, description, image URLs, and neighborhood.
• Strict JSON schema validation: always outputs snake_case keys.
• Handles proxy rotation, CAPTCHAs, and anti-bot protections using Bright Data’s MCP stack.
• Easily extendable for more data fields and custom sources.

🔒 Security Best Practices

• Store all API keys and credentials securely in your .env file.
• Always validate and sanitize extracted data before use.
• Respect robots.txt and website terms of service.