Provide audio notifications to users (macOS only)
AI agents invoke notify to trigger actions in Reviewer MCP. 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.
This tool executes a system-level operation on macOS to produce audio output, which is an external side effect beyond simple data retrieval or storage. It interacts with the OS notification/audio subsystem, making it Execute category. Misuse could be used for harassment or disruptive alerts, but blast radius is limited, hence medium severity.
From the tool's definition 'Provide audio notifications to users (macOS only)' — triggers an external system operation (audio playback/OS notification) on the host machine
Attacks that exploit this kind of access
Provide audio notifications to users (macOS only). It is categorised as a Execute tool in the Reviewer MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Reviewer MCP server in PolicyLayer and add a rule for notify: 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 Reviewer MCP. Nothing to install.
notify 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 notify 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 notify. 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.
notify is provided by the Reviewer MCP server (jaggederest/mcp_reviewer). 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 →