Live, verifiable data for robots and embodied agents.
An autonomous robot or embodied agent needs to know the world state it is acting in - is the GPS jammed in this region, is there an earthquake or storm front nearby, where is that satellite right now. Dynamic Feed hands your stack that live state as 94 keyless tools, and every datapoint arrives signed with source, observed-at, and freshness metadata for inspection.
Advisory evidence — not a safety system. Each verdict is a signed snapshot carrying observation times for public data: NOT a safety certification, safety system or safety function, and not a guarantee that acting is safe. Your control stack owns every action and applies its own safety logic; a valid signature establishes signed-byte integrity and attribution, not independent time, truth or the safety of a decision.
The call
One keyless POST pulls GPS-interference risk, recent quakes and a live satellite track together - each result carries its own signature and provenance. Point a robot or embodied agent at the MCP endpoint instead and the same tools appear at runtime:
curl -s -X POST https://dynamicfeed.ai/v1/batch \
-H 'Content-Type: application/json' \
-d '{"calls":[
{"tool":"gps_interference","args":{"region":"baltic"}},
{"tool":"earthquakes","args":{"min_magnitude":5.0}},
{"tool":"satellite_position","args":{"name":"ISS"}}
]}'Sample response
{ "mcpServers": { "dynamic-feed": { "url": "https://dynamicfeed.ai/mcp" } } }Why live data
Robots and embodied agents act on the physical world, so the data feeding their decisions has to be both fresh and auditable after the fact. Dynamic Feed gives you a neutral live-data layer over MCP (Streamable HTTP) and REST: GPS-interference maps, earthquakes, weather and marine forecasts, hazard and disaster feeds, satellite positions and more, all keyless. Every datapoint carries provenance (source, observed-at, freshness) and an Ed25519 signature. The signature establishes integrity and attribution under the published key, so altered bytes fail verification; it does not independently prove time, truth, accuracy, or safety. Your control stack still owns the safety decision; Dynamic Feed gives it citeable inputs carrying their source-reported times and a signed record of exactly what it was told.
Use it for
- Gate an outdoor robot or drone mission on live GPS-interference risk for its operating region
- Add hazard awareness - earthquakes, storms, marine conditions - to a navigation or planning loop
- Track a satellite or the ISS position at runtime for an aerial or orbital workflow
- Keep a signed, provenance-stamped log of the world-state your robot acted on for audit and incident review
FAQ
Do I need an API key?
No. The MCP endpoint and the POST /v1/batch REST endpoint are both keyless and free for fair use. Point your robot or agent at the URL and start calling tools.
What does the Ed25519 signature actually prove?
It establishes integrity and attribution under the published key, so altered bytes fail verification. It does not independently prove time, truth, accuracy, or safety - your control and safety logic still owns that decision.
How do I tell how fresh a datapoint is?
Every datapoint is provenance-stamped with its source and an observed-at timestamp plus a freshness indicator, so your agent can reject anything too stale for a real-time decision before it acts on it.