complex_search
AI agents call complex_search to retrieve information from Everything Search without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries local files using the Everything search engine with complex filtering (date, size, media type, document type). File search operations are read-only with no side effects—they query data without modifying, deleting, or executing code. The empty tool description prevents higher confidence, but the server context strongly indicates a read-only search function.
From the tool's definition The server enables 'search local files' and the tool name 'complex_search' indicates it performs file search operations.
Attacks that exploit this kind of access
complex_search. It is categorised as a Read tool in the Everything Search MCP Server, which means it retrieves data without modifying state.
Register the Everything Search MCP server in PolicyLayer and add a rule for complex_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 Everything Search. Nothing to install.
complex_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 complex_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 complex_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.
complex_search is provided by the Everything Search MCP server (miloira/everything-search-mcp). 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 →