SQL 式查詢單一資料集(SELECT-only,limit 上限 400)。
AI agents call query_rows to retrieve information from Healthcare OpenData MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool permits only SQL SELECT queries on healthcare open datasets with an enforced row limit, making it a pure read operation with no side effects. The SELECT-only restriction and explicit limit prevent any destructive, write, or execute capabilities. As part of a healthcare open data system, the blast radius of misuse is limited to unauthorized data retrieval from already-public datasets, warranting low severity.
From the tool's definition Tool description states 'SELECT-only' with a 400-row limit, explicitly restricting to read-only SQL queries with no modification or deletion capabilities.
Attacks that exploit this kind of access
SQL 式查詢單一資料集(SELECT-only,limit 上限 400)。. It is categorised as a Read tool in the Healthcare OpenData MCP MCP Server, which means it retrieves data without modifying state.
Register the Healthcare OpenData MCP server in PolicyLayer and add a rule for query_rows: 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 Healthcare OpenData MCP. Nothing to install.
query_rows 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 query_rows 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 query_rows. 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.
query_rows is provided by the Healthcare OpenData MCP server (trionnemesis/healthcare-opendata-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 →