A Python pipeline for MLS-format real estate data with AI agent support via MCP Protocol — enabling investment screening, automated valuation, and market analysis workflows.
PropData is a Python library providing programmatic, MLS-format access to real estate listing data from Realtor.com — active listings, sold comparables, rental market data, and pending transactions. It outputs structured data as pandas DataFrames, Pydantic models, CSV, Excel, or JSON, with a native MCP (Model Context Protocol) server that enables AI agents to query real estate data in natural language as part of automated investment analysis workflows.
Real estate investors, analysts, and PropTech companies need programmatic access to MLS-quality listing data — but traditional MLS data access costs $500–5,000 per month and requires REALTOR® membership in most markets. Automated valuation models, investment screening tools, and market trend analysis platforms all depend on current, structured property data that is freely accessible to consumers through web interfaces but architecturally inaccessible to programmatic workflows. The gap between what a human can see on Realtor.com and what a data pipeline can consume has historically required expensive licensing to bridge.
PropData targets the public-facing Realtor.com data layer and normalizes it into MLS-standard field schemas — the same field definitions REALTORS® use in their professional systems. Pydantic models provide type safety for every listing attribute: beds, baths, lot size, list price, days on market, price reductions, and geographic coordinates. Concurrent request handling enables fast bulk retrieval across large markets. The MCP Protocol integration is the forward-looking architectural decision: real estate AI agents built on Claude, GPT-4, or Gemini can call PropData as a tool, enabling natural language queries — “Find 3-bedroom homes sold in the last 90 days in ZIP 30303 under $500K” — that return structured data directly into the agent’s reasoning loop.
Tell us about your problem. We'll tell you honestly how we'd approach it — and whether we're the right team.