name: Device voice description: Read Ouroboros replies aloud using the voices available on your device. version: 1.0.0 type: extension entry: plugin.py plugin_api: "2.0" runtime: python3 os: any permissions: [page_module] when_to_use: The owner wants an Ouroboros reply read aloud. model_experience: what_model_sees: No new tools or messages; reading is an interface action. token_effect: No model tokens for reading or while idle. contributes:
provides: tts label: Device voice runtime: client destination: device render: {kind: host, entry: ui/tts.js}
Device voice
Enabled with the bundled skills. “Read aloud” reads a reply; “Stop” ends it. Local voices appear first. The voice menu says “On this device” only when the browser marks that voice local; other voices say “Online voice” and may send text to the operating system's speech service. Voice availability and languages depend on this device. No model is called and no model tokens are used.