calls_send_to_meet
ActiveTool of io.github.saloprj/dialogbrain
declared in 1.17.0
Dispatch a workspace AI agent into an active Google Meet call. The agent joins as a participant — it can hear the conversation, respond via TTS, see the shared screen (when vision is enabled on the agent), and answer questions about what's on screen. Use when the operator wants to delegate live meeting attendance to an agent (notes, Q&A, summarization, real-time support). The Meet URL must be in canonical 3-4-3 form, e.g. https://meet.google.com/abc-defg-hij. Lookup-redirect URLs are not supported — operator must use the share-link form.
Parameters schema
{
"type": "object",
"required": [
"meet_url",
"agent_id"
],
"properties": {
"agent_id": {
"type": "integer",
"description": "ID of an active agent in this workspace. Get it from agents.list. Any active agent can be dispatched — a voice trigger is NOT required (the runner attaches the agent you name directly)."
},
"meet_url": {
"type": "string",
"maximum": 120,
"description": "Canonical Google Meet URL — must match https://meet.google.com/<3 letters>-<4 letters>-<3 letters>, e.g. https://meet.google.com/abc-defg-hij. lookup/ redirects are NOT supported."
},
"vision_mode": {
"enum": [
"off",
"on_demand",
"continuous_0_3fps"
],
"type": "string",
"description": "Screen-share capture mode. 'off' = no vision (default), 'on_demand' = the agent can call the vision_query tool for fine-detail reads, 'continuous_0_3fps' = the bot captures the screen at 1 fps with phash dedupe and the executor splices the latest scene-change into each agent turn as ambient low-detail context. OMIT to use 'off' (the default)."
},
"instructions": {
"type": "string",
"description": "What the agent should do once it joins — its task brief, e.g. 'greet everyone and present the overview deck' or 'take notes and answer questions about the roadmap'. Woven into the agent's system prompt for the session. OMIT for a generic listening agent. Do NOT set with translation_mode — it is ignored there (translator mode needs no task brief)."
},
"app_languages": {
"type": "array",
"items": {
"type": "string"
},
"description": "Extra target languages translated for the APP ONLY (shown as subtitles + listenable audio in the inbox), never spoken into the Meet call. Operators switch between them in the UI. ISO codes, e.g. ['de', 'fr']. Max 4. Only with translation_mode=true. OMIT for a single spoken language."
},
"comeback_phrase": {
"type": "string",
"description": "Optional phrase to trigger return to original language during translation (e.g. 'Back to English'). Only used with translation_mode=true."
},
"target_language": {
"type": "string",
"description": "PRIMARY target language spoken INTO the Meet call when translation_mode=true (e.g. 'en', 'es', 'fr'). OMIT to not use translation."
},
"translation_mode": {
"type": "boolean",
"description": "If true, pre-arm translator mode (auto-target language) on join with zero verbal trigger. Forces the original-speaker audio track to exist. Requires target_language. OMIT (default false) for normal dispatch. This is the COMPLETE way to dispatch a live translator — do NOT also pass start_immediately or instructions (they are ignored here, and would otherwise start an agentic chat session instead of translating)."
},
"start_immediately": {
"type": "boolean",
"description": "If true, the agent starts talking as soon as it joins — it greets everyone and begins the task in `instructions` without waiting for someone to say a wake-word. OMIT (default false) to stay silent until addressed. MUTUALLY EXCLUSIVE with translation_mode — do NOT set both; translator mode is already hands-free and this is ignored (setting it would break translation)."
}
}
}Parent server
io.github.saloprj/dialogbrain
https://github.com/saloprj/dialogbrain-mcp
1/7 registries