Get list of counties from USGS Real-Time Flood Impacts API
AI agents call get_counties to retrieve information from USGS Water MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a straightforward data retrieval tool that queries a list of counties from a reference API. It has no side effects, cannot modify data, and poses minimal security risk. It falls clearly into the Read category as it simply fetches static reference information.
From the tool's definition The tool description indicates it 'Get list of counties', which is a retrieval operation. The USGS Water MCP server provides 'access to real-time water data' and allows users to 'fetch' measurements and retrieve reference data.
Attacks that exploit this kind of access
Get list of counties from USGS Real-Time Flood Impacts API. It is categorised as a Read tool in the USGS Water MCP MCP Server, which means it retrieves data without modifying state.
Register the USGS Water MCP server in PolicyLayer and add a rule for get_counties: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches USGS Water MCP. Nothing to install.
get_counties is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_counties rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for get_counties. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
get_counties is provided by the USGS Water MCP server (pgiffy/usgs-water-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →