AI agents call knitbrain_classify_task 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 tool analyzes and categorizes input (task classification) and returns structured metadata (tier, phases, plan-mode signal) for downstream consumption. It is purely analytical with no side effects—characteristic of Read operations. The 'follow the returned plan' instruction indicates the tool provides guidance/intelligence rather than directly executing actions.
From the tool's definition Tool classifies a task into tiers and returns plan-mode signals. The description uses only read/retrieval verbs: 'classify' (analyze/categorize existing data) and 'follow the returned plan' (consume output).
Attacks that exploit this kind of access
Classify a task into a tier (inquiry/trivial/standard/complex) with phases + plan-mode signal. Follow the returned plan. 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_classify_task: 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_classify_task 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_classify_task 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_classify_task. 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_classify_task 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 →