Dharmik2905/Custom_MCP_Neighbourhood

Custom_MCP_Neighbourhood

A comprehensive Model Context Protocol server that transforms AI assistants into intelligent neighborhood analysts by fetching and aggregating real-world location data from multiple APIs.

Purpose

The Neighborhood Intelligence MCP Server enables AI models to provide data-driven insights about any location by automatically gathering and synthesizing information from over 10 different data sources. Whether you're relocating, investing in real estate, or simply exploring a new area, this server transforms a simple address query into a comprehensive neighborhood analysis.

Why Neighborhood Intelligence Matters

The Real Cost of Poor Location Decisions The Relocation Crisis: Americans move an average of 11.7 times in their lifetime, yet 44% express regret about their last housing decision within the first year. The primary reason? Inadequate neighborhood research before committing.

Financial Impact:

The average cost of relocating (including moving, deposits, and transaction fees) ranges from $8,000 to $15,000 Property values can vary by 30-40% within a 2-mile radius in major metropolitan areas A poor neighborhood choice can cost homeowners $50,000-$150,000 in lost equity over a 5-year period

Quality of Life Statistics:

• Residents in walkable neighborhoods report 13% higher life satisfaction scores
• Crime rates can vary 10x between adjacent ZIP codes
• Commute times over 45 minutes are associated with 40% higher stress levels and increased divorce rates
• Air quality differences within a single city can reduce life expectancy by 1-3 years

Key Capabilities

• Smart Context Assembly – Automatically determines what data is relevant for each query
• Multi-Source Integration – Aggregates weather, crime, housing, walkability, and more
• AI-Powered Evaluation – Uses Claude 3.5 Sonnet to synthesize raw data into actionable insights
• Comparative Analysis – Evaluates and compares multiple neighborhoods side-by-side

🌐 Data Sources & APIs

Quick Start

Prerequisites

• Python 3.8+
• pip package manager
• Valid API keys (see Configuration)

Installation

#Clone the repository
git clone <your-repo-url>
cd neighborhood-intelligence-mcp

# Install dependencies
pip install mcp requests openai

Configuration

Create a .env file or export environment variables:

Required for AI evaluation

export OPENROUTER_API_KEY="sk-or-v1-..."

Required for functionality

export GOOGLE_MAPS_API_KEY="AIza..." export WALKSCORE_API_KEY="..." export RAPIDAPI_KEY="..." export AIR_QUALITY_API_KEY="..."

Optional fallback key

export ATTOM_API_KEY="..."

Running Server

Standalone mode

python server.py

Or integrate with Claude Desktop

Add to your MCP settings configuration

`

Example Prompts

When integrated with Claude Desktop, try these queries:

• "Evaluate the neighborhood around 123 Main Street, Austin, TX for a family with young children"
• "Compare the walkability of Downtown Dallas vs Uptown Dallas"
• "What's the air quality and crime rate near Central Park, New York?"
• "I'm moving to Houston for work at the Texas Medical Center. Find me a safe, affordable neighborhood within 30 minutes."

Available Tools

Architecture

┌─────────────────────────────────────────────────────────────────┐ │ 🏗️ ARCHITECTURE │ └─────────────────────────────────────────────────────────────────┘

                      👤 User Query
                            │
                            ↓
                  ┌─────────────────────┐
                  │  🏗️ MCP Server       │
                  │    (server.py)       │
                  └─────────────────────┘
                            │
                            ↓
    ┌───────────────────────────────────────────────┐
    │     ⚙️  Context Assembly Engine               │
    │  ┌─────────────────────────────────────────┐  │
    │  │  1. 📍 Geocode Address                  │  │
    │  │  2. 🎯 Determine Needed Data Sources    │  │
    │  │  3. ⚡ Fetch Data in Parallel           │  │
    │  │  4. 🛡️  Handle API Failures Gracefully  │  │
    │  └─────────────────────────────────────────┘  │
    └───────────────────────────────────────────────┘
                            │
                            ↓
    ┌───────────────────────────────────────────────┐
    │          📊 Data Sources (10+ APIs)           │
    ├───────────────────────────────────────────────┤
    │  ☀️  Weather  │  🚨 Crime    │  🏠 Housing    │
    │  🏫 Schools   │  🚇 Transit  │  🍽️  Dining    │
    │  💼 Demographics │ 🏥 Health │  🎭 Culture   │
    │  🌳 Parks     │  📈 Trends   │  ... More     │
    └───────────────────────────────────────────────┘
                            │
                            ↓
    ┌───────────────────────────────────────────────┐
    │         🤖 AI Evaluation Layer                │
    │                                               │
    │    Claude 3.5 Sonnet synthesizes raw data    │
    │         into actionable insights              │
    │                                               │
    │  • Identifies patterns                        │
    │  • Contextualizes statistics                  │
    │  • Generates recommendations                  │
    └───────────────────────────────────────────────┘
                            │
                            ↓
            ┌───────────────────────────┐
            │    ✨ Output              │
            │                           │
            │  📄 Comprehensive         │
            │  Neighborhood Report      │
            └───────────────────────────┘

`

Unique Features

Intelligent Fallbacks

• Predictive housing prices if Zillow unavailable
• OSM walkability fallback when WalkScore fails
• Synthetic safety index when no crime data exists

Multi-Source Walkability

• Combines Google Places, WalkScore, and OpenStreetMap density
• Provides a robust blended metric

Real Crime Data (2024)

• Powered by jgentes-crime-data
• Provides total incidents, trend analysis, and severity scoring

Comparative Analysis Mode

• demo.py evaluates multiple neighborhoods
• Uses Claude 3.5 Sonnet for natural-language comparative reports

Sample Output

` Running AI Comparative Analysis via Claude 3.5 Sonnet...

Comparative AI Summary:

Here's a structured comparison of Downtown Austin and North Dallas:

OVERALL SCORES: Downtown Austin: 8.2/10 North Dallas: 7.0/10

BEST SUITED FOR: Downtown Austin:

• Young professionals (25-35)
• Tech workers
• Singles/couples without children
• Nightlife enthusiasts
• Urban lifestyle seekers
• Higher income individuals

North Dallas:

• Young families
• Education-focused households
• Mid-career professionals
• Suburban-minded urbanites
• School-age children
• Those seeking balance of urban/suburban

CORE METRICS COMPARISON: Housing Costs:

• Downtown Austin: Higher ($1,850 rent, $550k median home)
• North Dallas: Moderate ($1,600 rent, $375k median home)

Amenities:

• Downtown Austin: Strong entertainment, dining, cultural venues
• North Dallas: Strong educational facilities, family-oriented services

Safety:

• Both areas rate 7/10 for safety (based on available data)

ONE-LINE RECOMMENDATIONS: Downtown Austin: "Ideal for young professionals seeking vibrant urban lifestyle with strong career opportunities, despite higher living costs."

North Dallas: "Solid choice for families prioritizing education and suburban qualities while maintaining urban conveniences."

Best Overall Choice: Downtown Austin (Due to higher overall score, stronger amenities, and better alignment with current urban development trends.)

Searching for best neighbourhood mode

Input: check the college_station_test.py. Demonstration of the Neighborhood Intelligence MCP Server focused on College Station, TX.

Evaluates multiple neighborhoods and provides individual and comparative AI analysis.

Output:

Comparative AI Summary:

Here's your structured comparison of College Station neighborhoods:

1. NEIGHBORHOOD OVERVIEWS

• Southwood Valley: Established family-friendly area with strong educational infrastructure and moderate housing costs, though car-dependent
• Wolf Pen Creek District: [Limited data available]
• Pebble Creek: Upscale suburban development farther from town center, offering quiet living and good schools but requiring longer commutes

2. COMPARATIVE INSIGHTS Southwood Valley:

• Best for: Young families, university staff
• Strengths: School proximity, 10-min campus commute
• Weaknesses: Poor walkability

Pebble Creek:

• Best for: Mid-career professionals, established families
• Strengths: Quieter environment, safety
• Weaknesses: 16-min commute, very car-dependent

Wolf Pen Creek:

• [Insufficient data for comparison]

3. RANKED BY OVERALL LIVABILITY 1️⃣ Southwood Valley (7.2/10)

• Better location
• Shorter commute
• More amenities nearby
• Same safety score as Pebble Creek

2️⃣ Pebble Creek (7.2/10)

• More suburban feel
• Longer commute
• Fewer nearby amenities
• Equal safety rating

3️⃣ Wolf Pen Creek District

• Unable to rank due to insufficient data

Key Differentiators:

• Location: Southwood Valley is more centrally located
• Commute: Southwood Valley (10 mins) vs Pebble Creek (16 mins)
• Amenities: Southwood Valley has better access to services
• Schools: Both areas have excellent schools
• Safety: Both rate 7/10
• Housing Costs: Similar in both areas ($285,000 median)

Best Overall Neighborhood: Southwood Valley Reasoning: While both rated neighborhoods score 7.2/10, Southwood Valley edges out Pebble Creek due to its more convenient location, shorter commute times, and better access to amenities while maintaining the same safety rating and housing costs. It offers the best balance of family-friendly features and practical convenience for College Station residents.

College Station Comparative Demo Complete.

Contributing

This project was developed for the Build-Your-Own-MCP Challenge @ TAMUDATATHON 2025. We welcome contributions! Please submit pull requests or open issues for:

New data source integrations Enhanced evaluation algorithms Performance optimizations Documentation improvements

License

MIT License – Free to use, fork, and modify.

Acknowledgments

Anthropic for MCP and Claude AI OpenStreetMap for geospatial data RapidAPI for housing and crime APIs Google Maps and WalkScore for local metrics OpenRouter for model orchestration

Support

For issues, questions, or feature requests:

Open an issue on GitHub Contact the development team Check the documentation wiki