/introduction
Introduction
Provenance-tagged geospatial data for any US coordinate.
What you get
- POST /v1/ask — a planner LLM picks the relevant fields, fetches them in parallel, and a synthesizer LLM writes the answer with inline citations. Returns the answer plus the full citation list.
- POST /v1/fetch — the developer escape hatch. Pass the fields or preset you want; get raw per-field values with provenance and a partial_failures array for anything we couldn’t fetch.
- POST /v1/geocode — start from a street address instead of a coordinate. Returns the coordinate and how it was derived, so you can tell a rooftop match from a street-interpolated estimate before you trust it.
- POST /v1/lookup — turn a messy address, coordinate, or APN into canonical join keys: a coordinate, resolved address, and — when it’s safe to — a parcel with area/zoning/sale history, plus county/tract, hazard, and market context in one call.
- MCP server — exposes these operations as MCP tools so Claude Desktop, Cursor, and custom MCP agents call Mireye natively.
Why it’s different
Federal sources only
Per-field citations
MCP-native distribution
Honest pattern
Coverage
- Geography: United States only — lower 48, Alaska, Hawaii, and US territories. Bounds: lat ∈ [18, 72], lng ∈ [-180, -65]. Out-of-bounds requests return 400 coord_out_of_bounds.
- Fields: 296 across 7 layers (terrain, land cover, built environment, utilities, parcels, climate, hazards). Full catalog at GET /v1/meta/fields.
- Presets: 15 use-case bundles (terrain, flood_risk, wildfire_underwrite, land_cover, site_selection, building_lookup, points_of_interest, utilities, boundaries, solar_siting, wind_siting, storage_siting, data_center_siting, grid_interconnect, natural_hazard).