Search for specific content within a document.
AI agents call search_document to retrieve information from DocNav-MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Search operations are inherently Read category tools - they query and retrieve data without side effects. The tool cannot delete, modify, execute code, or perform financial transactions. The blast radius of misuse is minimal; an agent could only retrieve information the user already has access to. Low severity reflects the absence of any capability to cause irreversible changes or execute arbitrary operations.
From the tool's definition Tool description states 'Search for specific content within a document' - this is a query/search operation that retrieves information without modifying or executing anything.
Attacks that exploit this kind of access
Search for specific content within a document. It is categorised as a Read tool in the DocNav-MCP MCP Server, which means it retrieves data without modifying state.
Register the DocNav- MCP server in PolicyLayer and add a rule for search_document: 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 DocNav-MCP. Nothing to install.
search_document 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_document 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_document. 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_document is provided by the DocNav- MCP server (shenyimings/docnav-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 →