AI agents call stape_gateway_resource to retrieve information from Stape without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool is primarily a retrieval/query mechanism for gateway configurations and resources. The verb 'retrieving' and lack of any write, destructive, or execution keywords classify this as a Read operation. The blast radius is limited since gateway configuration retrieval does not inherently enable harmful actions, though the severity could increase if sensitive credentials or architectural details are exposed.
From the tool's definition Tool name and description indicate 'retrieving gateway-related resources and configurations' with no modification, deletion, or execution capability mentioned.
Attacks that exploit this kind of access
Tool for retrieving gateway-related resources and configurations. Use the. It is categorised as a Read tool in the Stape MCP Server, which means it retrieves data without modifying state.
Register the Stape MCP server in PolicyLayer and add a rule for stape_gateway_resource: 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 Stape. Nothing to install.
stape_gateway_resource 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 stape_gateway_resource 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 stape_gateway_resource. 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.
stape_gateway_resource is provided by the Stape MCP server (stape-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 →