AI agents call advanced-search to retrieve information from Ripgrep without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs text searches across files—a read-only operation with no side effects on data. Even though it supports advanced options, ripgrep itself cannot modify, delete, or execute code; it only retrieves matching text. No reversible or irreversible modifications are possible. The blast radius of misuse is minimal: an agent could search for sensitive patterns but cannot act on them without other tools.
From the tool's definition Tool is part of ripgrep-mcp server described as providing 'text searches across files on your system'. The tool name 'advanced-search' and description 'Advanced search with ripgrep with more options' indicate a search capability.
Attacks that exploit this kind of access
Advanced search with ripgrep with more options. It is categorised as a Read tool in the Ripgrep MCP Server, which means it retrieves data without modifying state.
Register the Ripgrep MCP server in PolicyLayer and add a rule for advanced-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 Ripgrep. Nothing to install.
advanced-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 advanced-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 advanced-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.
advanced-search is provided by the Ripgrep MCP server (shk1447/ripgrep-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 →