create_policy
AI agents use create_policy to create or update resources in PolicyGuard — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your PolicyGuard environment.
This tool creates new policies that govern AI agent behavior and enforce security rules. Policy creation is a reversible Write operation (policies can be updated or deleted), but carries high severity because malicious or misconfigured policies could disable legitimate security controls, bypass access restrictions, or weaken compliance monitoring across the organization.
From the tool's definition Tool name 'create_policy' combined with server context (PolicyGuard provides policy-based access control and governance) indicates creation of access control policies.
Attacks that exploit this kind of access
create_policy. It is categorised as a Write tool in the PolicyGuard MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the PolicyGuard MCP server in PolicyLayer and add a rule for create_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 PolicyGuard. Nothing to install.
create_policy is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the create_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 create_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.
create_policy is provided by the PolicyGuard MCP server (prateekkumar1709/policyguard). 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 →