Search PubMed by keyword, MeSH term, or author. Returns PMIDs and titles.
AI agents call search_pubmed to retrieve information from Pubmed Search without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves publicly available bibliographic metadata from PubMed without side effects. It is a classic search/query operation that poses minimal risk—an AI agent cannot misuse it to damage data, execute code, or cause financial harm. The blast radius is negligible.
From the tool's definition Tool 'search_pubmed' returns search results (PMIDs and titles) in response to queries; no modification, deletion, or execution of external code is described.
Attacks that exploit this kind of access
Search PubMed by keyword, MeSH term, or author. Returns PMIDs and titles. It is categorised as a Read tool in the Pubmed Search MCP Server, which means it retrieves data without modifying state.
Register the Pubmed Search MCP server in PolicyLayer and add a rule for search_pubmed: 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 Search. Nothing to install.
search_pubmed 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 search_pubmed 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 search_pubmed. 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.
search_pubmed is provided by the Pubmed Search MCP server (pubspro/pubmed-search). 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 →