For developers & AI
Toronto property data for your AI agent.
The same City records behind PropertyMonitorTO, exposed over the Model Context Protocol. Ask Claude, Cursor, or any MCP client about a Toronto address in plain language — answered from real permits, orders, and neighbourhood data, with the source cited. Free, open, and stateless (no accounts, nothing stored).
Connect
Streamable-HTTP endpoint — point any MCP client at it:
https://mcp.propertymonitor.to/mcp
Claude Desktop / claude.ai and ChatGPT (Developer Mode, paid plans) have no one-click link yet — add a custom connector in settings and paste the URL above.
Claude Code
claude mcp add --transport http propertymonitor https://mcp.propertymonitor.to/mcp
Cursor / generic mcp.json
{
"mcpServers": {
"propertymonitor": { "url": "https://mcp.propertymonitor.to/mcp" }
}
}What your agent can do
Find
- Search any Toronto address → a propertyId for the other tools
This property
- Overview — permits, open orders, and investigations on file
- Building permit history
- Open orders & violations
- Bylaw investigations
The building
- RentSafeTO building report card
- Short-term-rental registration
The neighbourhood
- Development applications nearby
- Crime rates vs the city average
- 2021 Census profile — income, tenure, demographics
- Industrial pollution sources nearby (federal NPRI)
Sources & privacy
City of Toronto open data (building permits, orders & violations, bylaw investigations, development applications, RentSafeTO evaluations, short-term-rental registrations, neighbourhood crime rates, and the 2021 Census) plus the federal National Pollutant Release Inventory — all under Open Government Licences. Every tool returns its dataset, licence, and vintage so answers can be cited. The server is stateless: it reads public records per request and stores nothing about the caller. Google-derived tools (rooftop solar, air quality) are intentionally not exposed — their terms don’t permit redistribution.