AI agents call list_safety_policy to retrieve information from Unlimited without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and returns existing safety policy settings. It performs no write, delete, execute, or financial operations. Confidence is high because the verb 'list/return' and the passive retrieval semantics are unambiguous. Severity is low because reading safety configuration data poses minimal risk — it does not expose sensitive secrets, execute code, or trigger external actions.
From the tool's definition Tool name is 'list_safety_policy' and description states it 'Return[s] the current safety configuration, allowed_roots and deny_paths' — a read-only retrieval of configuration data with no modification or side effects.
Attacks that exploit this kind of access
Return the current safety configuration, allowed_roots and deny_paths. It is categorised as a Read tool in the Unlimited MCP Server, which means it retrieves data without modifying state.
Register the Unlimited MCP server in PolicyLayer and add a rule for list_safety_policy: 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 Unlimited. Nothing to install.
list_safety_policy 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 list_safety_policy 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 list_safety_policy. 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.
list_safety_policy is provided by the Unlimited MCP server (triumsebas/unlimited-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 →