// SCAN REPORT

Voice Mode

25 tools. 11 can modify or destroy data without limits.

3 destructive tools with no built-in limits. Policy required.

Last updated: 10 Jun 2026

11 can modify or destroy data

14 read-only

25 tools total

Community server · catalogue entry verified 10/06/2026

How to control Voice Mode ↓

TOOL MAP

What Voice Mode exposes to your agents

Read (14) Write / Execute (6) Destructive / Financial (3)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Voice Mode tools

Destructive · High kokoro_uninstall This tool performs irreversible deletion and removal operations. Destructive · Low voice_statistics_reset The tool irreversibly clears all tracked metrics and statistics. Destructive · High whisper_uninstall This tool performs irreversible deletion of software installations, service configurations, downloaded models, and user data (logs, transcriptions). Execute · Low config_reload This tool triggers an operational action (reloading config and clearing caches) rather than simply reading or writing data.

11 of Voice Mode's 25 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Voice Mode

PolicyLayer is an MCP gateway — it sits between your AI agents and Voice Mode, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "kokoro_uninstall": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "refresh_provider_registry": {
    "limits": [
      {
        "counter": "refresh_provider_registry_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "check_audio_dependencies": {
    "limits": [
      {
        "counter": "check_audio_dependencies_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Voice Mode — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON VOICE MODE →

Free to start. No card required.

FULL CATALOGUE

All 25 Voice Mode tools

DESTRUCTIVE 3 tools

Destructive kokoro_uninstall Uninstall kokoro-fastapi and optionally remove models and data. This tool will: 1. Stop any runni Destructive voice_statistics_reset Reset all voice conversation statistics and start a new session. Clears all tracked metrics and reset Destructive whisper_uninstall Uninstall whisper.cpp and optionally remove models and data. This tool will: 1. Stop any running

EXECUTE 4 tools

Execute config_reload Reload configuration from .voicemode.env files and clear all caches. This tool reloads configuration Execute kokoro_install kokoro_install Execute whisper_install whisper_install Execute whisper_model_install whisper_model_install

WRITE 2 tools

Write refresh_provider_registry Manually refresh health checks for voice provider endpoints. Useful when a service has been started/s Write update_config Update a configuration value in the voicemode.env file. Args: key: The configuration key to u

READ 14 tools

Read check_audio_dependencies Check system audio dependencies and provide installation guidance. This tool checks for required syst Read check_audio_devices List available audio input and output devices Read get_provider_details Get detailed information about a specific provider endpoint. Args: base_url: The base URL of Read get_whisper_version Get Whisper installation and version information. Read list_config_keys List all available configuration keys with their descriptions. Returns: A formatted list of a Read list_tts_voices List available TTS voices for different providers. Args: provider: Optional provider name ('o Read show_config_files Show which .voicemode.env files are being used for configuration. This shows the current configuratio Read voice_mode_info Get diagnostic information about the voice-mode installation. Shows version, configuration, and provi Read voice_registry Get the current voice provider registry showing all discovered endpoints. Returns a formatted view of all Read voice_statistics Display live statistics dashboard for voice conversation performance. Shows current session statistic Read voice_statistics_recent Show recent voice conversation interactions with timing details. Args: limit: Maximum number Read voice_statistics_summary Get a concise summary of voice conversation performance metrics. Returns key performance indicators f Read voice_status Check the status of all voice services including TTS, STT, LiveKit, and audio devices. IMPORTANT: Onl Read voice_statistics_export Export detailed voice conversation statistics as JSON data. Exports all tracked metrics and calculate

OTHER 2 tools

Other converse converse Other service service

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Voice Mode

Can an AI agent delete data through the Voice Mode MCP server? +

Yes. The Voice Mode server exposes 3 destructive tools including kokoro_uninstall, voice_statistics_reset, whisper_uninstall. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Voice Mode? +

The Voice Mode server has 2 write tools including refresh_provider_registry, update_config. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Voice Mode.

How many tools does the Voice Mode MCP server expose? +

25 tools across 3 categories: Destructive, Read, Write. 14 are read-only. 11 can modify, create, or delete data.

How do I enforce a policy on Voice Mode? +

Register the Voice Mode MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Voice Mode tool call.

Deterministic rules across all 25 Voice Mode tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON VOICE MODE →

Free to start. No card required.

25 Voice Mode tools catalogued and risk-classified — across an index of 43,000+ MCP servers.