AI agents call search-poems to retrieve information from Pplog without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries data from the pplog service without producing side effects. The search functionality is informational only—it does not create, modify, or delete poems, nor does it execute code or trigger external operations. It falls clearly into the Read category with low severity since misuse would only expose existing poem data rather than cause harm or data loss.
From the tool's definition Tool name 'search-poems' and description 'Searches for poems in pplog with advanced query syntax support' indicate a query/search operation with no modification, deletion, or execution of arbitrary code.
Attacks that exploit this kind of access
Searches for poems in pplog with advanced query syntax support. It is categorised as a Read tool in the Pplog MCP Server, which means it retrieves data without modifying state.
Register the Pplog MCP server in PolicyLayer and add a rule for search-poems: 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 Pplog. Nothing to install.
search-poems 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-poems 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-poems. 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-poems is provided by the Pplog MCP server (oci:ghcr.io/esaio/pplog-mcp-server:0.4.1). 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 →