AI agents call kg_maintain to retrieve information from Predicate without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Even though kg_maintain only reads data, uncontrolled read access leaks sensitive information and racks up API costs — an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Attacks that exploit this kind of access
Archive stale low-confidence ABox triples; emit a MaintenanceRun event. It is categorised as a Read tool in the Predicate MCP Server, which means it retrieves data without modifying state.
Register the Predicate MCP server in PolicyLayer and add a rule for kg_maintain: 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 Predicate. Nothing to install.
kg_maintain 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 kg_maintain 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 kg_maintain. 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.
kg_maintain is provided by the Predicate MCP server (nordicagents/predicate). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
kg_maintain is one line of Predicate's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →