The Model Context Protocol (MCP) is an open standard that enables AI models to securely connect to external data sources and tools. Houski's MCP integration allows AI models to query Canadian property data directly, enabling powerful real estate analysis and insights.
Through MCP, AI models can access all Houski API endpoints with natural language queries, making property data analysis more intuitive and powerful than ever before.
Houski's MCP server works with any MCP-compatible client. Popular options include:
Client | Description | Best For |
---|---|---|
Claude Desktop | Anthropic's official desktop application with built-in MCP support | General users, real estate professionals |
Claude Web Interface | Anthropic's web-based interface with MCP capabilities | Cross-platform users, browser-based workflows |
Claude messages API | Programmatic access through Anthropic's API with MCP tools | Developers, custom applications |
Cursor | Code editor with AI capabilities and MCP integration | Developers, data analysts, technical users |
Custom MCP Clients | Build your own MCP client using the open standard | Enterprise integrations, specialized workflows |
With MCP, you can ask natural language questions and the AI model will intelligently utilize the Houski API to answer them:
Construction planning:
"Where in Calgary would be an optimal location to build a new hospital?"
Urban planning insights:
"Where in Calgary would be optimal to build a new hospital based on population density, income levels, and transit access?"
Climate risk analysis:
"Show me properties in Vancouver with high flood risk scores but low earthquake risk - I need this for insurance underwriting"
Investment arbitrage:
"Find undervalued properties in Toronto where the list price is 20% below the AI estimated value and they have permits for secondary suites"
Gentrification tracking:
"Which neighborhoods in Montreal have seen the biggest increase in renovation permits and new construction in the last 2 years?"
Lifestyle matching:
"I'm a 28-year-old software engineer who bikes to work. Find condos under $600k in areas with high education levels, good internet scores, and strong bike infrastructure"
Market timing strategy:
"Show me properties that have had multiple price decreases in the last 6 months and are now priced below their 2023 assessment values"
Houski's MCP server mirrors the entire functionality of the Houski API, providing access to all property data tools. Your AI model of choice can call these tools using natural language queries.
Tool Name | Description | Use Cases |
---|---|---|
get_data_for_properties | Retrieve detailed property data with filtering, sorting, and expansion options | Market analysis, property research, investment screening |
aggregate_properties | Get statistical aggregations (median, mean, percentiles) for market analysis | Market comparisons, pricing analysis, demographic insights |
geocode_properties | Find properties around specific coordinates with radius-based search | Location-based searches, proximity analysis, CMA reports |
get_locations | Retrieve hierarchical geographic data (countries, provinces, cities, communities) | Market research, location discovery, geographic analysis |
get_map_data | Generate map visualization data including pins, clusters, and heatmaps | Spatial analysis, market visualization, trend mapping |
predict_property_values | Generate historical property value estimates using machine learning models | Trend analysis, model backtesting, historical research |
fuzzy_search_address_tool | Search for specific properties using flexible address matching | Property identification, address lookup, unit searches |
validate_api_key | Validate API key authentication and test connectivity | Setup verification, troubleshooting, key validation |
search_api_documentation | Access Houski API documentation for learning and reference | Documentation lookup, learning, troubleshooting |
MCP tool calls are billed according to our standard API pricing. Each tool call corresponds to an API request and is charged based on the data returned.
See our pricing page for detailed cost information.
Need help with MCP setup or have questions? We're here to help: