read_files
AI agents call read_files to retrieve information from File Search Tool without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and reads file contents without modification, deletion, or execution. It is a pure read operation with no side effects. The empty tool description is compensated by the clear server-level documentation and sibling tool names (all read-only: list, search, get) that establish the read-only nature of this MCP server.
From the tool's definition Tool name 'read_files' combined with server description stating it enables 'reading text, PDF, and DOCX files' and sibling tools focused on 'directory listing, regex-based file name and content searches'.
Attacks that exploit this kind of access
read_files. It is categorised as a Read tool in the File Search Tool MCP Server, which means it retrieves data without modifying state.
Register the File Search Tool MCP server in PolicyLayer and add a rule for read_files: 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 File Search Tool. Nothing to install.
read_files 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 read_files 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 read_files. 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.
read_files is provided by the File Search Tool MCP server (rjtpp/mcp-server-file-search-tool). 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 →