AI agents call check-permission to retrieve information from Spicedb without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool performs a read-only authorization query against SpiceDB. It retrieves permission information but does not create, modify, delete, or execute any external operations. The function is purely informational, determining whether a permission relationship exists.
From the tool's definition Tool name is 'check-permission' and description states it 'Checks whether a specific subject has a particular permission on a resource' - this is a query operation that retrieves authorization status without modifying any data.
Attacks that exploit this kind of access
Checks whether a specific subject has a particular permission on a resource. This is the core authorization check function of SpiceDB - determining. It is categorised as a Read tool in the Spicedb MCP Server, which means it retrieves data without modifying state.
Register the Spicedb MCP server in PolicyLayer and add a rule for check-permission: 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 Spicedb. Nothing to install.
check-permission 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 check-permission 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 check-permission. 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.
check-permission is provided by the Spicedb MCP server (samkim/spicedb-mcp-server). 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 →