AI agents call fhir.search to retrieve information from FhirMCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and retrieves healthcare data (FHIR resources) without modifying or deleting anything. While severity is elevated to 'high' due to the sensitive nature of PHI (Protected Health Information) in FHIR servers — unauthorized searches could expose patient records, medications, diagnoses, and other confidential medical data — the tool itself performs only read operations.
From the tool's definition Tool description states it "Search[es] FHIR resources with parameters, pagination, and field selection" — a retrieval operation with no modification.
Attacks that exploit this kind of access
Search FHIR resources with parameters, pagination, and field selection. It is categorised as a Read tool in the FhirMCP MCP Server, which means it retrieves data without modifying state.
Register the Fhir MCP server in PolicyLayer and add a rule for fhir.search: 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 FhirMCP. Nothing to install.
fhir.search 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 fhir.search 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 fhir.search. 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.
fhir.search is provided by the Fhir MCP server (xsovx/fhir-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 →