AI agents call get_bookshelf to retrieve information from Weread without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and displays user bookshelf data with statistics and categories. It performs no writes, deletes, executions, or financial operations. The sole risk is information disclosure of the user's reading list, which is low severity since it's the user's own data being queried and the impacts are limited to privacy/data exposure rather than destructive or consequential actions.
From the tool's definition Tool name is 'get_bookshelf' and description states it 'Get all books in the user's bookshelf' — a query/retrieval operation with no modification capability.
Attacks that exploit this kind of access
Get all books in the user's bookshelf with comprehensive statistics and categorization information. It is categorised as a Read tool in the Weread MCP Server, which means it retrieves data without modifying state.
Register the Weread MCP server in PolicyLayer and add a rule for get_bookshelf: 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 Weread. Nothing to install.
get_bookshelf 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_bookshelf 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_bookshelf. 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_bookshelf is provided by the Weread MCP server (mcp-server-weread). 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 →