AI agents call knitbrain_read 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 retrieves and queries file data with optimization features (skeleton view, recall hash for pagination), but produces no side effects. The description emphasizes it as a read operation designed to return information rather than modify or execute. Classified as Read category with low severity due to limited blast radius from an AI agent misusing file retrieval alone.
From the tool's definition Tool name 'knitbrain_read' and description explicitly states 'Read a project file' with retrieval functionality ('returns a structure-preserving skeleton'). No modification, deletion, or execution capability described.
Attacks that exploit this kind of access
Read a project file OPTIMIZED: returns a structure-preserving skeleton (signatures/schema kept, bulk elided) + a ⟨recall:hash⟩ to page in the exact original. Use INSTEAD of the host. 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_read: 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_read 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_read 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_read. 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_read 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 →