← All templates
Agent ToolSupport & research bots

Grounded Location Q&A

A complete Agent Skill — copy it and drop it into your agent’s skills folder as grounded-location-qa/SKILL.md and it can run this recipe on its own.

Give any support or research bot a tool that answers plain-English questions about a location — with citations, not hallucination.

Single QueryStreaming Chat

How it works

  1. Wire POST /v1/ask as a tool call: it takes a location (lat/lng or an address, never both) and a natural-language question.
  2. Mireye's planner picks the right catalog fields, fetches them, and a synthesizer model writes a cited prose answer, plus a confidence level and a citations list grouped by source.
  3. For a chat UI, swap in POST /v1/ask/stream (SSE) to stream the answer token-by-token instead of waiting for the full response — same fields, delivered incrementally.

Example response

lat
46.6
lng
-93.7
question
What is the wildfire risk at this location and what fuel is on the ground?
answer
Wildfire risk at this location (46.6, -93.7) is LOW. The land cover is classified as Grass/Forb/Herb with 0.0% tree canopy cover, meaning there is no forested fuel structure present. The terrain is essentially flat at 0.05 degrees slope, which provides no topographic amplification of fire spread.
confidence
medium
citations
source
USFS_LCMS
source_url
https://developers.google.com/earth-engine/datasets/catalog/USFS_GTAC_LCMS_v2024-10
fields
lcms_class
confidence
high
source
USGS_EPQS
source_url
https://epqs.nationalmap.gov/v1/json
fields
elevation
confidence
high
fields_used
elevation
lcms_class
ndvi_change_5y
ndvi_current
slope_degrees
tree_canopy_pct
resolved_location
lat
46.6
lng
-93.7
source
coordinate

Get a Mireye API key at mireye.com/account.