Search indexed files using natural language queries. Finds files containing content semantically similar to the query. When to use this tool: - Find documentation, guides, or explanations about specific topics - Locate code files implementing certain functionality or patterns - Discover configura...
AI agents call search to retrieve information from Directory Indexer without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool purely retrieves and queries indexed content without any side effects. It does not create, modify, delete, or execute anything. The severity is low because unauthorized access to file metadata through semantic search poses minimal risk compared to other categories, though sensitivity depends on indexed content.
From the tool's definition Tool description states it "Search indexed files using natural language queries" and "Finds files containing content semantically similar to the query." The use cases specify retrieval operations: "Find documentation", "Locate code files", "Discover…
Documented attack patterns abuse exactly the kind of access search gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Directory Indexer, and nothing reaches the server without passing your rules. This is the rule we recommend for search:
{
"version": "1",
"default": "deny",
"tools": {
"search": {}
}
}search 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 indexed files using natural language queries. Finds files containing content semantically similar to the query. When to use this tool: - Find documentation, guides, or explanations about specific topics - Locate code files implementing certain functionality or patterns - Discover configuration files, scripts, or settings related to a topic - Search for files covering specific concepts or technologies How it works: - Converts query to vector embedding using semantic similarity - Searches all indexed file chunks for relevant content - Groups results by file and calculates average relevance scores - Returns files ranked by relevance score Examples: -. It is categorised as a Read tool in the Directory Indexer MCP Server, which means it retrieves data without modifying state.
Register the Directory Indexer MCP server in PolicyLayer and add a rule for 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 Directory Indexer. Nothing to install.
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 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 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.
search is provided by the Directory Indexer MCP server (peteretelej/directory-indexer). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Directory Indexer, 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.
7 Directory Indexer tools catalogued and risk-classified — across an index of 43,000+ MCP servers.