AI agents call lookup_icd10 to retrieve information from Medterms without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves medical terminology data (ICD-10 codes and descriptions) without creating, modifying, deleting, or executing any operations. It is a read-only query against a medical code database. The blast radius if misused is minimal—an AI agent cannot cause harm through retrieving diagnostic codes. No side effects occur.
From the tool's definition Tool description states 'Search ICD-10-CM codes by diagnosis name or look up a specific code. Returns code, full description, and category.' This is purely a lookup/search operation with no modification or execution of external systems.
Attacks that exploit this kind of access
Search ICD-10-CM codes by diagnosis name or look up a specific code. Returns code, full description, and category. Essential for clinical documentation agents. It is categorised as a Read tool in the Medterms MCP Server, which means it retrieves data without modifying state.
Register the Medterms MCP server in PolicyLayer and add a rule for lookup_icd10: 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 Medterms. Nothing to install.
lookup_icd10 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 lookup_icd10 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 lookup_icd10. 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.
lookup_icd10 is provided by the Medterms MCP server (pubspro/medterms-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 →