Trouve l
AI agents call geocodage_inverse to retrieve information from French Admin without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Reverse geocoding is a read-only query operation that retrieves location or address information based on geographic coordinates. There are no side effects, no data modification, and no code execution. The truncated description prevents full certainty, but the tool name and context (French administration helper) strongly suggest this is a simple lookup/retrieval service.
From the tool's definition Tool name 'geocodage_inverse' (reverse geocoding) and truncated description 'Trouve l' (finds/retrieves) indicate data retrieval. No modification, deletion, or execution language present.
Attacks that exploit this kind of access
Trouve l. It is categorised as a Read tool in the French Admin MCP Server, which means it retrieves data without modifying state.
Register the French Admin MCP server in PolicyLayer and add a rule for geocodage_inverse: 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 French Admin. Nothing to install.
geocodage_inverse 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 geocodage_inverse 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 geocodage_inverse. 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.
geocodage_inverse is provided by the French Admin MCP server (mcp-tools-lab/french-admin-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 →