Data Collection
unlistedhomes.com
Fetches property details and searches for properties on Unlisted.
ENDPOINT 1
https://api.unlistedinc.com/mcp
MCP server metadata
- Name
- unlisted
- Version
- 1.0.0
Unlisted is a central source of information for all residential properties in the USA -- every property not just those that are for sale and listed on the MLS etc. Use these tools together to search for and retrieve property information: ## Typical workflow 1. **AutocompleteTool** — Has two purposes, (1) resolve a place name (city, neighborhood, county, school district, etc.) to a boundary_id, OR (2) resolve an address (e.g. 123 Some St Somewhere AZ 12345) to a property_id Call this first whenever the user mentions a location by name or a single specific address. 2. **PropertySearchTool** — Search for properties using filters. Always pass the boundary_id from step 1 unless the user explicitly wants a nationwide search. Combine boundary_id with other filters (beds, baths, price range, perks, etc.) as needed. When few filters are used, you can still do the search but encourage the user to provide more filters to narrow results. 3. **PropertyDetailsTool** — Fetch full details for a single property by ID, for example after the user selects one from search results. ## Key rules - Use boundary_id to scope searches geographically — never search without one unless the user is explicitly asking for nationwide results, or a appropriate boundary was not found, in which case a "geometry" can be generated and used instead. - boundary_id and geometry can be combined: results satisfy both constraints (intersection). - When generating geometry do you best to use an appropriate amount of decimal places, and let the user know when an approximation is made that may impact the results. - lot_area_min/lot_area_max are in **acres**, not square feet. - do not refer to our "property profiles" as "listings". A given property may be "listed" for sale somewhere else, and we may link to those listings, but we do not host listings. - your understanding of boundaries is important for interfacing with these tools, but users don't need to know about "boundaries". Talk as specifically about areas of interest as the user does. - do not use unnecessary details when describing your use of these tools. For example: instead of "I’m pulling the Wisconsin state boundary first, then I’ll sort by most recently owner-updated so the top result is the freshest property profile in the state." you can just say "Stand by while I find the freshest property profile in the state!"
Known tools 3
autocomplete_toolAutocomplete a search term to find geographic areas (cities, counties, neighborhoods, school districts, states, etc.
Inferred read-onlyCONNECT WITH APPROVAL
Client installation
Review this server and its permissions before adding it. Secret placeholders must be set locally.
Codex
~/.codex/config.toml
[mcp_servers.unlisted]
url = "https://api.unlistedinc.com/mcp"
enabled = true
Claude Code
.mcp.json
{
"mcpServers": {
"unlisted": {
"type": "http",
"url": "https://api.unlistedinc.com/mcp"
}
}
}
Claude Desktop
Settings → Connectors → Add custom connector
Name: unlisted
Remote MCP URL: https://api.unlistedinc.com/mcp
Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.
Cursor
.cursor/mcp.json
{
"mcpServers": {
"unlisted": {
"url": "https://api.unlistedinc.com/mcp"
}
}
}
Visual Studio Code
.vscode/mcp.json
Add to Visual Studio Code{
"servers": {
"unlisted": {
"type": "http",
"url": "https://api.unlistedinc.com/mcp"
}
}
}
Generic MCP
Client-specific MCP configuration
{
"name": "unlisted",
"transport": "streamable-http",
"url": "https://api.unlistedinc.com/mcp"
}
MCP Inspector
Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.
TRUST AND VERIFICATION EVIDENCE
Loading Trust v2 evidence…
Checking the associated registrable domain. The BuiltWith key remains server-side.
Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.