AI agents call search_medrxiv_key_words to retrieve information from medRxiv-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 preprint metadata from the medRxiv repository based on keyword queries. No side effects, data modification, execution, or destructive operations are indicated. The tool is purely informational/retrieval in nature. Confidence is slightly reduced due to empty tool description, but the server context and naming pattern strongly indicate read-only search functionality.
From the tool's definition Tool name 'search_medrxiv_key_words' indicates a search operation. Server description states it enables AI assistants to 'search and access medRxiv papers' with no mention of modification or deletion capabilities.
Documented attack patterns abuse exactly the kind of access search_medrxiv_key_words gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and medRxiv-MCP-Server, and nothing reaches the server without passing your rules. This is the rule we recommend for search_medrxiv_key_words:
{
"version": "1",
"default": "deny",
"tools": {
"search_medrxiv_key_words": {}
}
}search_medrxiv_key_words 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_medrxiv_key_words. It is categorised as a Read tool in the medRxiv-MCP-Server MCP Server, which means it retrieves data without modifying state.
Register the medRxiv-MCP-Server MCP server in PolicyLayer and add a rule for search_medrxiv_key_words: 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 medRxiv-MCP-Server. Nothing to install.
search_medrxiv_key_words 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_medrxiv_key_words 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_medrxiv_key_words. 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_medrxiv_key_words is provided by the medRxiv-MCP-Server MCP server (jackkuo666/medrxiv-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from medRxiv-MCP-Server, 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.
3 medRxiv-MCP-Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.