Search for semantically similar document chunks using local embeddings and vector similarity
AI agents call semanticSearchLocal to retrieve information from Medical 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 and ranks existing documents based on semantic similarity using pre-computed embeddings. It has no side effects: it does not create, modify, delete, or execute arbitrary operations. The blast radius of misuse is minimal—an attacker could retrieve sensitive medical documents they shouldn't access, but cannot modify or destroy data.
From the tool's definition Tool performs 'Search for semantically similar document chunks' — a query operation with no modification, deletion, or execution of external code. The verb 'search' and the passive nature of vector similarity matching indicate data retrieval only.
Attacks that exploit this kind of access
Search for semantically similar document chunks using local embeddings and vector similarity. It is categorised as a Read tool in the Medical MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Medical MCP Server MCP server in PolicyLayer and add a rule for semanticSearchLocal: 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 Medical MCP Server. Nothing to install.
semanticSearchLocal 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 semanticSearchLocal 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 semanticSearchLocal. 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.
semanticSearchLocal is provided by the Medical MCP Server MCP server (kalyankumarkonduru/mcp-server). 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 →