Set the role for an agent. Roles control which tools the agent can call.
AI agents use cortex_set_role to create or update resources in Cortex MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Cortex MCP environment.
This tool modifies agent configuration by assigning a role, which in turn governs access to other tools. While it is a write/update operation (reversible in principle), the blast radius is high because misconfiguring or escalating an agent's role could grant it access to destructive or financial tools it otherwise wouldn't have, effectively enabling privilege escalation.
From the tool's definition Set the role for an agent. Roles control which tools the agent can call.
Attacks that exploit this kind of access
Set the role for an agent. Roles control which tools the agent can call. It is categorised as a Write tool in the Cortex MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Cortex MCP server in PolicyLayer and add a rule for cortex_set_role: 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 Cortex MCP. Nothing to install.
cortex_set_role 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 cortex_set_role 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 cortex_set_role. 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.
cortex_set_role is provided by the Cortex MCP server (neuralnexustech/cortex-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 →