pico_search
AI agents call pico_search to retrieve information from PubMed Enhanced Search without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
PICO search is a querying mechanism for retrieving academic papers based on structured criteria. It retrieves data with no side effects—it neither modifies, deletes, executes code, nor moves money. The empty description slightly reduces confidence, but context from server purpose and sibling tools confirms this is a read-only search function.
From the tool's definition Tool is named 'pico_search' and resides on a PubMed search server described as enabling 'search and retrieval of academic papers.' PICO (Patient/Population, Intervention, Comparison, Outcome) is a structured evidence search methodology.
Attacks that exploit this kind of access
pico_search. It is categorised as a Read tool in the PubMed Enhanced Search MCP Server, which means it retrieves data without modifying state.
Register the PubMed Enhanced Search MCP server in PolicyLayer and add a rule for pico_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 PubMed Enhanced Search. Nothing to install.
pico_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 pico_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 pico_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.
pico_search is provided by the PubMed Enhanced Search MCP server (leescot/pubmed-mcp-smithery). 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 →