AI agents invoke test_notification to trigger actions in Wardrowbe. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
The tool triggers an external operation (sending a notification) whose effect depends on the provided notification setting argument. This is an Execute-category action as it initiates an external side effect (pushing a notification), but it is not destructive, financial, or a simple data read/write. Severity is medium because misuse could spam users with notifications or probe notification infrastructure.
From the tool's definition "Fire a test notification through the given notification setting"
Attacks that exploit this kind of access
Fire a test notification through the given notification setting. It is categorised as a Execute tool in the Wardrowbe MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Wardrowbe MCP server in PolicyLayer and add a rule for test_notification: 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 Wardrowbe. Nothing to install.
test_notification is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the test_notification 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 test_notification. 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.
test_notification is provided by the Wardrowbe MCP server (saya6k/mcp-wardrowbe). 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 →