AI agents call get_books to retrieve information from MCPServer without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves metadata (document names) from object storage. It performs a read-only list operation on a bucket, returning base64-encoded names. There is no capability to modify, delete, or execute code. The blast radius is minimal—an AI agent calling this tool can only discover what documents exist in the storage, which is informational only.
From the tool's definition Tool description states "Get list of documents" - a retrieval operation with no modification or deletion. The phrase "get list" indicates querying/listing data from MinIO storage with no side effects.
Attacks that exploit this kind of access
Get list of documents from MinIO storage under /files prefix in documents bucket. Names are base64 encoded. It is categorised as a Read tool in the MCPServer MCP Server, which means it retrieves data without modifying state.
Register the MCPServer MCP server in PolicyLayer and add a rule for get_books: 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 MCPServer. Nothing to install.
get_books 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 get_books 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 get_books. 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.
get_books is provided by the MCPServer MCP server (ldilba/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 →