Model Context Protocol (MCP) servers provide a machine-consumable interface that
allows Large Language Models (LLMs) to interact
directly
with our
geospatial web APIs. They expose structured tools—defined by schemas, parameter types, and
clear
contracts—
so that models can request Earth intelligence data reliably and without custom glue code.
The Amentum MCP gateway exposes our production data services at
https://mcp.amentum.io/mcp/{service}/. Available service names are
ocean, atmosphere, avrad, geomag,
and gravity. The gateway is deployed on AWS behind HTTPS and forwards your
API-Key header to the underlying Amentum APIs without storing credentials.
To use the MCP server, create an account in the developer portal, generate an API key, and
subscribe to the services your agent needs. The same API key is used for both the web APIs
and MCP access, so your MCP client only needs to send the key as an API-Key
header.
Example: list the tools exposed by the Ocean MCP server with the FastMCP Python client.
import asyncio
import os
from fastmcp import Client
from fastmcp.client.transports import StreamableHttpTransport
async def main():
transport = StreamableHttpTransport(
url="https://mcp.amentum.io/mcp/ocean/",
headers={"API-Key": os.environ["AMENTUM_API_KEY"]},
)
client = Client(transport, timeout=120, init_timeout=120)
async with client:
tools = await client.list_tools()
print(f"Discovered {len(tools)} MCP tools:")
for tool in tools[:5]:
print(f"- {tool.name}")
asyncio.run(main())
Expected output:
Discovered 6 MCP tools:
- app_api_ausseabed_endpoints_AusSeabed_Bathymetry_point
- app_api_bathymetry_endpoints_GEBCO_Bathymetry_point
- app_api_nemo_endpoints_MFWAM_point
- app_api_nemo_endpoints_NEMO_point
- app_api_nemo_endpoints_NEMO_Phys_point
Exact tool names can change as API schemas evolve, but the workflow is stable: connect to
the service endpoint, discover available tools, then let your agent choose and call the
tool whose schema matches the user request.