AI 检索：用户提问，返回与该问题相关的医学证据列表
AI agents call ai_search to retrieve information from KnowS 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 searches medical evidence based on user queries. It is a read-only operation that queries existing data and presents results without creating, modifying, deleting, or executing external operations. The low severity reflects that misuse would at worst retrieve irrelevant or incorrect medical information for analysis, without direct patient harm or system compromise.
From the tool's definition Tool description states it takes user questions and 'returns a list of medical evidence related to that question' (ai_search → user query → medical evidence list). The verb 'returns' indicates data retrieval with no modification.
Attacks that exploit this kind of access
AI 检索：用户提问，返回与该问题相关的医学证据列表. It is categorised as a Read tool in the KnowS MCP Server MCP Server, which means it retrieves data without modifying state.
Register the KnowS MCP Server MCP server in PolicyLayer and add a rule for ai_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 KnowS MCP Server. Nothing to install.
ai_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 ai_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 ai_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.
ai_search is provided by the KnowS MCP Server MCP server (pancrepal-xiaoyibao/mcp-knows-ai). 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 →