Live weather, hazard, and environmental data for vision and field robots.
A robot operating outdoors needs to know the conditions around it right now - rain in the next hour, an active severe-weather alert, smoke or haze degrading its cameras, or a region where GPS is being jammed. Dynamic Feed gives it that from one keyless endpoint, with the source and timestamp attached to every reading.
The call
One keyless POST returns the near-term forecast, any active severe-weather alerts, the local air quality, and a measured GPS-interference reading - each item signed and stamped with its source and observed-at time:
curl -s -X POST https://dynamicfeed.ai/v1/batch \
-H "Content-Type: application/json" \
-d '{"calls":[
{"tool":"weather_forecast","args":{"city":"Sydney","days":2}},
{"tool":"weather_alerts","args":{"area":"FL"}},
{"tool":"air_quality","args":{"city":"Delhi"}},
{"tool":"gps_interference","args":{"region":"baltic"}}
]}'Sample response
{ "mcpServers": { "dynamic-feed": { "url": "https://dynamicfeed.ai/mcp" } } }Why live data
A robot that plans paths or trusts its sensors from stale or unsourced data is flying blind. Dynamic Feed attaches a source, an observed-at time, and a freshness state to every reading, plus an Ed25519 signature so you can log it and later prove the exact bytes you acted on. Note what the signature does and does not mean: it is proof-of-existence and tamper-evidence - it proves Dynamic Feed reported this value at this time and that nobody altered it in transit. It is NOT a claim that the value is true, accurate, or safe to act on. Weather and hazard feeds come from upstream sources (NWS/NOAA, Open-Meteo, and others); the gps_interference reading is derived by Dynamic Feed from public ADS-B navigation-integrity data. Use the signature for an audit trail and the freshness stamp for your own go/no-go logic - don't treat a signature as a safety certification.
Use it for
- Gate an outdoor mission on active severe-weather alerts and the next-hour forecast
- Flag camera-degrading conditions (smoke, haze, fog) before a vision robot acts on a frame
- Warn a GNSS-dependent rover or drone when it enters a region with measured GPS interference
- Log a signed, source-stamped record of the conditions a robot acted on for incident review
FAQ
Do I need an API key?
No. The /v1/batch REST endpoint and the MCP server at https://dynamicfeed.ai/mcp are both keyless and free for fair use. Point your robot's runtime or agent at the URL and call the tools.
Which environmental and hazard tools are here?
weather_forecast and current_weather for conditions, weather_alerts for active US severe-weather warnings (NWS/NOAA), air_quality for PM2.5/PM10 and AQI, plus earthquakes, weather_alerts, air_quality and gps_interference for hazard awareness. gps_interference is a Dynamic Feed reading derived from live ADS-B navigation integrity, useful for GNSS-reliant robots. These sit alongside 87 tools across 19 verticals and 70 sources.
Does the signature mean the data is accurate or safe to act on?
No. The Ed25519 signature is proof-of-existence and tamper-evidence: it proves Dynamic Feed reported this value at this observed-at time and that the bytes were not altered. It is not a truth, accuracy, or safety guarantee. Use it for an audit trail and run your own go/no-go logic on the freshness stamp.