高德 POI 关键字搜索（POI 2.0 /v5/place/text）
AI agents call search_poi to retrieve information from Amap MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves POI (Points of Interest) information based on keyword searches. It is a read-only query operation that returns geographic and location data without side effects, state changes, or destructive actions. The blast radius of misuse is minimal—an agent could only retrieve unwanted location data or exhaust API quotas, not cause persistent damage or financial harm.
From the tool's definition Tool name 'search_poi' with description indicating POI keyword search via Amap API. POI (Points of Interest) searches are queries that retrieve geographic data without modifying or deleting anything.
Attacks that exploit this kind of access
高德 POI 关键字搜索（POI 2.0 /v5/place/text）. It is categorised as a Read tool in the Amap MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Amap MCP Server MCP server in PolicyLayer and add a rule for search_poi: 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 Amap MCP Server. Nothing to install.
search_poi 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 search_poi 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 search_poi. 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.
search_poi is provided by the Amap MCP Server MCP server (qiao101660/gaode_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 →