Return per-vendor and per-model aggregation of all recorded measurements.
AI agents call metrics_summary to retrieve information from Librarian MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves and aggregates benchmark metrics data that has already been recorded. It performs a read-only operation with no side effects, no data modification, and no capability to execute code or trigger external operations. The aggregation is computed over existing measurements, making this a straightforward data retrieval function.
From the tool's definition Tool description: 'Return per-vendor and per-model aggregation of all recorded measurements.' This is a retrieval/query operation that aggregates pre-recorded data without modifying, executing, or deleting anything.
Attacks that exploit this kind of access
Return per-vendor and per-model aggregation of all recorded measurements. It is categorised as a Read tool in the Librarian MCP MCP Server, which means it retrieves data without modifying state.
Register the Librarian MCP server in PolicyLayer and add a rule for metrics_summary: 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 Librarian MCP. Nothing to install.
metrics_summary 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 metrics_summary 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 metrics_summary. 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.
metrics_summary is provided by the Librarian MCP server (liana-banyan/librarian-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 →