AI agents use assign-client-role-to-user to create or update resources in Keycloak — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Keycloak environment.
This tool modifies user permissions by assigning client roles, which is a reversible Write operation. However, the severity is high because incorrect role assignments can grant unintended access to sensitive resources or services. The Keycloak context (IAM system) amplifies the impact of misuse, as compromised role assignments affect authentication and authorization across dependent systems.
From the tool's definition Tool description states it 'Assign a client role to a user' — a modification operation that creates or updates a role assignment.
Attacks that exploit this kind of access
Assign a client role to a user. It is categorised as a Write tool in the Keycloak MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Keycloak MCP server in PolicyLayer and add a rule for assign-client-role-to-user: 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 Keycloak. Nothing to install.
assign-client-role-to-user 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 assign-client-role-to-user 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 assign-client-role-to-user. 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.
assign-client-role-to-user is provided by the Keycloak MCP server (keycloak-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 →