AI agents call aim_memory_unlink to permanently remove resources in Mcp Knowledge Graph — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Unlinking memories removes the relationship between them, which is not easily reversible without knowing the original link structure. While it doesn't delete the memory nodes themselves, destroying the graph edges can permanently alter the knowledge graph topology. This qualifies as Destructive rather than Write because relational structure, once removed, may not be recoverable.
From the tool's definition 'Remove links between memories' — explicitly removes/unlinks relationships between memory nodes
Documented attack patterns abuse exactly the kind of access aim_memory_unlink gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mcp Knowledge Graph, and nothing reaches the server without passing your rules. This is the rule we recommend for aim_memory_unlink:
{
"version": "1",
"default": "deny",
"hide": [
"aim_memory_unlink"
]
}aim_memory_unlink disappears from the agent's tool list entirely, and any attempt to call it is denied. The rest of the server keeps working.
Free to start. No card required.
Remove links between memories. It is categorised as a Destructive tool in the Mcp Knowledge Graph MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Mcp Knowledge Graph MCP server in PolicyLayer and add a rule for aim_memory_unlink: 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 Mcp Knowledge Graph. Nothing to install.
aim_memory_unlink is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the aim_memory_unlink 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 aim_memory_unlink. 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.
aim_memory_unlink is provided by the Mcp Knowledge Graph MCP server (shaneholloman/mcp-knowledge-graph). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 10 Mcp Knowledge Graph tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
10 Mcp Knowledge Graph tools catalogued and risk-classified — across an index of 42,500+ MCP servers.