Skip to main content
Glama

@pipeworx/nasa-firms

Satellite active-fire detections from NASA FIRMS — every pixel MODIS or VIIRS measured as actively burning, published within ~3 hours of satellite overpass.

Part of Pipeworx — an MCP gateway connecting AI agents to 1679+ live data sources.

Tools

  • firms_hotspots_area(bbox, source?, day_range?, date?, limit?) — detections inside a bounding box. bbox is west,south,east,north in degrees, or "world".

  • firms_hotspots_near(latitude, longitude, radius_km?, source?, day_range?, date?) — detections around a point, sorted nearest first with distance_km.

  • firms_data_availability(source?) — first and last date each sensor product covers. Call it before a historical query so an empty result can be read as "outside the archive" rather than "no fires".

Related MCP server: mcp-nasa-eonet

Auth

BYO only. FIRMS requires a MAP_KEY and publishes no keyless or demo path — verified 2026-09-17: every route answers Invalid MAP_KEY. (HTTP 400/401) without one. Pass it as _apiKey; the platform fallback env var is PLATFORM_NASA_FIRMS_KEY.

The key is free and self-service at https://firms.modaps.eosdis.nasa.gov/api/map_key/ — an email address, no approval step. Limit is 5,000 transactions per 10 minutes, and a multi-day query counts as several transactions.

Data sources

Things the next person would otherwise rediscover

  • acq_time is HHMM with leading zeros dropped. "52" means 00:52Z. A naive two-character slice reports it as 52:00 and gets the hour wrong for every pre-10:00 overpass.

  • Brightness column names differ by sensor. VIIRS reports bright_ti4 / bright_ti5; MODIS reports brightness / bright_t31. Both are mapped to brightness_k / bright_t31_k.

  • The /api/country/ and /api/countries/ routes are dead. FIRMS' own API index lists them as "Feature currently not available" and they answer Invalid API call. (HTTP 400). firms_hotspots_country was therefore NOT built. Do not emulate it with a country bounding box — a box around Chile contains Argentina, and the caller would get neighbouring fires labelled as the country they asked for. firms_hotspots_near is the honest shape.

  • day_range is capped at 5 by the upstream, and is counted FORWARD from date when one is given, backward from today when it is not.

  • Errors arrive as a bare English sentence, not JSON, at HTTP 400/401/403 — so the response body is the diagnosis and is surfaced verbatim.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "nasa-firms": {
      "url": "https://gateway.pipeworx.io/nasa-firms/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/nasa-firms/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1679+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/firms_hotspots_area \
  -H 'Content-Type: application/json' \
  -d '{"bbox":"-125,32,-114,42","source":"VIIRS_SNPP_NRT","day_range":1}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/firms_hotspots_area. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Standalone (no gateway account)

This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:

{
  "mcpServers": {
    "nasa-firms": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-nasa-firms"]
    }
  }
}

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-nasa-firms

It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call for only this pack's tools — none of the shared meta-tools the gateway connection above adds. Same source, same tools, no ask_pipeworx routing.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Nasa Firms data" })

The gateway picks the right tool and fills the arguments automatically.

More

License

MIT

Related MCP Connectors

Related MCP Servers

  • A
    license
    Not graded
    quality
    C
    maintenance
    Enables querying NASA's Earth Observatory Natural Event Tracker (EONET) for natural events, categories, sources, and visualization layers.
    1 npm
    MIT
  • F
    license
    Not graded
    quality
    B
    maintenance
    Near-real-time worldwide wildfire ignition data from NASA FIRMS, GOES and Meteosat MTG satellites plus AI-verified witness reports, served by kanari.io as a remote MCP server (no API key). Tools: active_fires, fire_archive_search, fire_details, wildfire_stats, firefighting_aircraft, earliness_cases. Open data, CC BY 4.0.
    1
    -
  • A
    license
    Not graded
    quality
    C
    maintenance
    Enables querying Sentinel-1/2, Landsat Collection 2, NAIP, and Copernicus DEM imagery by area and date, returning item IDs, footprints, cloud cover, and cloud-optimized GeoTIFF URLs.
    MIT