AI agents use relation_create to create or update resources in Mementos — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mementos environment.
This tool creates new relations in a memory system, which is a write operation that modifies data state but remains reversible (relations can be deleted via relation_delete or similar operations). The blast radius is medium because misuse could corrupt the agent's knowledge graph or create false semantic relationships, but the effects are confined to metadata and can be undone.
From the tool's definition Tool name 'relation_create' and description 'Create a relation between two entities' explicitly indicates data creation via a reversible operation.
Attacks that exploit this kind of access
Create a relation between two entities (uses, knows, depends_on, created_by, related_to, contradicts, part_of, implements). It is categorised as a Write tool in the Mementos MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mementos MCP server in PolicyLayer and add a rule for relation_create: 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 Mementos. Nothing to install.
relation_create is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the relation_create 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 relation_create. 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.
relation_create is provided by the Mementos MCP server (@hasna/mementos). 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 →