AI agents invoke notify_speak to trigger actions in Claude Voice Notify. 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 an external TTS engine to synthesize and play arbitrary audio output. It is not a simple data read/write — it triggers an auditory operation through an external provider (VOICEVOX, Google Cloud TTS, OpenAI, etc.), making it Execute. Misuse could include playing disruptive, misleading, or socially engineered audio, giving it medium severity.
From the tool's definition 「任意のテキストを音声で読み上げる」(reads arbitrary text aloud via voice), with configurable notification type, speed, and speaker — triggers an external TTS engine operation
Documented attack patterns abuse exactly the kind of access notify_speak gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Claude Voice Notify, and nothing reaches the server without passing your rules. This is the rule we recommend for notify_speak:
{
"version": "1",
"default": "deny",
"tools": {
"notify_speak": {
"limits": [
{
"counter": "notify_speak_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}notify_speak stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
任意のテキストを音声で読み上げる。通知タイプ・速度・話者を指定可能。. It is categorised as a Execute tool in the Claude Voice Notify MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Claude Voice Notify MCP server in PolicyLayer and add a rule for notify_speak: 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 Claude Voice Notify. Nothing to install.
notify_speak 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_speak 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_speak. 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_speak is provided by the Claude Voice Notify MCP server (kubouchiyuya/claude-voice-notify). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Claude Voice Notify, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
5 Claude Voice Notify tools catalogued and risk-classified — across an index of 43,000+ MCP servers.