Manage knowledge graph: read metadata, get stats, or delete graph
AI agents call manage_graph to permanently remove resources in Mnemosyne MCP — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
While the tool also supports non-destructive operations (read metadata, get stats), the most severe capability available is 'delete graph', which would irreversibly destroy an entire knowledge graph and all its contents. Per the rules, the most severe applicable category must be chosen, making this Destructive.
From the tool's definition 'or delete graph' — the tool explicitly supports deleting an entire knowledge graph, which is an irreversible destructive operation
Attacks that exploit this kind of access
Manage knowledge graph: read metadata, get stats, or delete graph. It is categorised as a Destructive tool in the Mnemosyne MCP MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Mnemosyne MCP server in PolicyLayer and add a rule for manage_graph: 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 Mnemosyne MCP. Nothing to install.
manage_graph 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 manage_graph 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 manage_graph. 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.
manage_graph is provided by the Mnemosyne MCP server (sophia-labs/mnemosyne-mcp). 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 →