An AI agent is only as good as the data it can reach. To prepare a flight, compute performance or check a procedure, it needs a source that is reliable, structured and traceable. That is what we are opening today: NOTAM Briefer and Pilot Briefer can now be used by any compatible AI agent, through the MCP protocol.
MCP in two sentences
The Model Context Protocol (MCP) is an open standard. It lets an AI agent such as Claude discover and call tools hosted on an external server, then reason over what comes back.
See it in action
In this video, a pilot asks Claude which NOTAMs are in force at Palo Alto (KPAO). Real question, real data.
What happens, step by step
- The question is asked in plain language.
- Claude loads the tools published by the NOTAM Briefer MCP server.
- Claude calls the tool “Airfield NOTAM briefing”.
- The server answers with structured data: every NOTAM carries its verification status, its position and how confident that position is.
- Claude writes the summary for a local VFR flight and cites the NOTAM numbers.
- Everything stays traceable: the raw request and response are one click away.
Try it
Both servers are public and need no authentication. In Claude, go to Settings, then Connectors, then “Add custom connector”, and enter the address:
- NOTAM Briefer:
https://notam.amenai.net/mcp - Pilot Briefer (weather):
https://briefer.amenai.net/mcp
NOTAM Briefer is not an official aeronautical information source. It does not replace a briefing from certified sources.
For your business
We started with our consumer apps. The same approach is coming to our enterprise products, starting with ELA: specialised agents that work from your own documentation, such as performance charts or torque tables, to run calculations with the sources attached. They can be called from the AI client your teams already use, and deployed on sovereign infrastructure.
Want to connect your agents to your data? Let’s talk.
