AI agents call search_core 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.
A search operation in an academic database context is a Read operation—it retrieves data from PubMed or similar research databases without modifying, deleting, or executing external code. Even though the description is empty, the tool name and server context provide sufficient evidence that this performs literature retrieval rather than modification or execution.
From the tool's definition Tool name 'search_core' combined with server context as 'research assistant MCP server, providing task-oriented literature search'. The name 'search_core' strongly suggests a search/query operation.
Documented attack patterns abuse exactly the kind of access search_core gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pubmed Search, and nothing reaches the server without passing your rules. This is the rule we recommend for search_core:
{
"version": "1",
"default": "deny",
"tools": {
"search_core": {}
}
}search_core is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
search_core. 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_core: 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_core 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_core 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_core. 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_core is provided by the Pubmed Search MCP server (u9401066/pubmed-search-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Pubmed Search, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
63 Pubmed Search tools catalogued and risk-classified — across an index of 43,000+ MCP servers.