AI agents call knitbrain_metrics to retrieve information from Knitbrain without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a metrics/telemetry retrieval tool that gathers and returns observability data about the knitbrain system's internal performance (compression efficiency, retrieval statistics). It performs no write operations, deletions, code execution, or financial transactions. The description explicitly indicates it retrieves counts and rates, making it a Read operation with minimal blast radius if misused by an AI agent.
From the tool's definition The tool retrieves 'compression telemetry: recall-store tier counts + per-kind retrieval rates' - it queries and reports metrics about system performance without modifying, deleting, or executing external operations.
Attacks that exploit this kind of access
Compression telemetry: recall-store tier counts + per-kind retrieval rates (TOIN self-tuning). It is categorised as a Read tool in the Knitbrain MCP Server, which means it retrieves data without modifying state.
Register the Knitbrain MCP server in PolicyLayer and add a rule for knitbrain_metrics: 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 Knitbrain. Nothing to install.
knitbrain_metrics 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 knitbrain_metrics 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 knitbrain_metrics. 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.
knitbrain_metrics is provided by the Knitbrain MCP server (pdgit12/knitbrain). 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 →