AI agents call graph_stats to retrieve information from Mementos without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and aggregates metadata about the knowledge graph structure without creating, modifying, or deleting any data. It is purely informational, analogous to a reporting or statistics-retrieval function. No side effects or irreversible actions are possible. Severity is low because misuse cannot harm the underlying data or cause operational damage.
From the tool's definition Tool retrieves graph statistics (counts, most-connected entities, orphan count, degree metrics). The description uses only query verbs: 'counts', 'most-connected', 'orphan count', 'average degree'—no modification or deletion operations.
Attacks that exploit this kind of access
Comprehensive knowledge graph statistics: entity/relation counts by type, most-connected entities, orphan count, average degree. It is categorised as a Read tool in the Mementos MCP Server, which means it retrieves data without modifying state.
Register the Mementos MCP server in PolicyLayer and add a rule for graph_stats: 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.
graph_stats 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 graph_stats 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 graph_stats. 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.
graph_stats 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 →