AI agents call query_concept to retrieve information from Engram 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 data from a knowledge graph without creating, modifying, or deleting any information. It is a pure read operation that returns related concepts and their strengths. No side effects or state changes occur. Blast radius is minimal since the worst outcome is retrieving irrelevant or sensitive information the AI already has access to.
From the tool's definition Tool name 'query_concept' and description 'Look up a concept in the knowledge graph. Returns related concepts and connection strengths.' indicate a lookup/retrieval operation with no modification of data.
Attacks that exploit this kind of access
Look up a concept in the knowledge graph. Returns related concepts and connection strengths. It is categorised as a Read tool in the Engram MCP Server, which means it retrieves data without modifying state.
Register the Engram MCP server in PolicyLayer and add a rule for query_concept: 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 Engram. Nothing to install.
query_concept 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 query_concept 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 query_concept. 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.
query_concept is provided by the Engram MCP server (tinydarkforge/engram). 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 →