Skip to main content

Documentation Index

Fetch the complete documentation index at: https://www.mireye.ai/llms.txt

Use this file to discover all available pages before exploring further.

The Mireye Earth MCP server provides two tools — ask and fetch. Both operate on US coordinates and return per-field provenance, but they serve different purposes: ask answers natural-language questions with cited prose, while fetch returns structured field values for programmatic use.

​
Tool: ask

Use ask when the caller has a specific question about a place — for example, “is this in a flood zone?” or “what’s the wildfire risk here?” The tool determines which fields are relevant, fetches them, and returns a cited answer in plain prose. 
ask(lat, lng, question)
​
lat
number
required
Latitude of the coordinate. Must be in [18, 72].
​
lng
number
required
Longitude of the coordinate. Must be in [-180, -65].
​
question
string
required
Natural-language question about the coordinate.
The tool returns the answer as cited prose along with per-source provenance: 
{
  "answer": "The elevation at 40.7128, -74.0060 is 13.15 meters above NAVD88...",
  "confidence": "HIGH",
  "citations": [
    {
      "source": "USGS_3DEP",
      "source_url": "https://www.usgs.gov/3d-elevation-program",
      "fields": ["elevation"],
      "fetched_at": "2026-05-24T22:14:01.882Z",
      "confidence": "HIGH"
    }
  ],
  "fields_used": ["elevation"]
}

​
Tool: fetch

Use fetch when you know which fields you need, or when you’re powering a workflow that consumes structured values — a scoring pipeline, a data export, or an underwriting calculation. Each field in the response includes its value, source, and confidence. 
fetch(lat, lng, fields?, preset?)
​
lat
number
required
Latitude of the coordinate. Must be in [18, 72].
​
lng
number
required
Longitude of the coordinate. Must be in [-180, -65].
​
fields
string[]
List of catalog field names to fetch. See Field discovery for how to look up available names.
​
preset
string
A named bundle of related fields. One of: terrain, flood_risk, wildfire_underwrite, land_cover, site_selection, building_lookup, utilities, boundaries.
The tool returns a fields map with one entry per requested field, and a partial_failures list for any fields that could not be fetched: 
{
  "fields": {
    "elevation": {
      "value": 13.15,
      "unit": "meters",
      "source": "USGS_3DEP",
      "source_url": "https://www.usgs.gov/3d-elevation-program",
      "confidence": "HIGH",
      "fetched_at": "2026-05-24T22:15:10.110Z",
      "ttl_seconds": 31536000,
      "notes": null
    }
  },
  "partial_failures": []
}

​
Which tool should I use?

SituationUse
Answering a question for a humanask
Feeding values into a scoring pipelinefetch
You know exactly which fields you needfetch with fields
You want a standard bundle of related fieldsfetch with preset
You want the agent to decide what’s relevantask

​
Field discovery

There is no list_fields MCP tool. To discover available fields and presets, call the meta endpoint directly: 
GET https://mireye-earth.fly.dev/v1/meta/fields
The endpoint sets a 1-hour Cache-Control and an ETag, so a long-lived agent can fetch it once at startup and reuse the result.