Skip to main content
miniti can POST a meeting payload as json to any url when a meeting is saved or when insights are regenerated. works with zapier, make, n8n, and any custom endpoint.

​
configure

  1. open settings (⌘,)
  2. go to the webhooks section
  3. paste a url into webhook url
  4. leave it empty to disable
miniti sends requests directly from your device. no backend sits in the middle.

​
events

eventfires when
meeting.savedyou save a meeting from a live / stopped session
meeting.updatedyou regenerate insights on an already-saved meeting

​
request

  • method: POST
  • headers: Content-Type: application/json
  • timeout: 10 seconds
  • retries: none (fire-and-forget)
  • signing: none (the webhook url is the shared secret - keep it private)
webhooks are unsigned. anyone with your webhook url can send payloads to it. use a hard-to-guess url path, or put a reverse proxy in front of your endpoint that checks a secret token from the url.

​
payload

{
  "event": "meeting.saved",
  "meeting": {
    "id": "3F9A2B1C-4D5E-4F6A-9B7C-0D1E2F3A4B5C",
    "title": "acme discovery call",
    "date": "2026-04-20T14:00:00Z",
    "end_time": "2026-04-20T14:32:17Z",
    "duration_seconds": 1937,
    "language": "en",
    "summary": "brief overview of the meeting...",
    "action_items": [
      "send follow-up email to jane",
      "schedule product demo next tuesday"
    ],
    "key_decisions": [
      "proceed with 30-day pilot"
    ],
    "topics": [
      "pricing",
      "onboarding",
      "integrations"
    ],
    "discussion_flow": [
      "introductions",
      "pain points",
      "product walkthrough",
      "next steps"
    ],
    "notes": "jane seems to own budget. mentioned competitor x.",
    "meddpicc": {
      "metrics": "targeting 40% lift in pipeline",
      "economic_buyer": "jane (vp ops)",
      "decision_criteria": "ease of onboarding, support sla",
      "decision_process": "pilot then exec review",
      "paper_process": "security + legal (2-week turnaround)",
      "identified_pain": "manual data entry across tools",
      "champion": "jane",
      "competition": "competitor x"
    },
    "training": {
      "talk_ratio_you": 0.42,
      "duration_minutes": 32.3,
      "speakers": [
        {
          "speaker": "you",
          "is_you": true,
          "word_count": 1843,
          "words_per_minute": 143.2,
          "fillers_per_minute": 2.1,
          "total_fillers": 68,
          "fillers": { "um": 24, "uh": 18, "like": 26 },
          "longest_monologue_words": 312,
          "questions_asked": 14,
          "avg_words_per_turn": 21.4
        }
      ]
    },
    "questions": [
      {
        "question": "what would make the pilot a clear success?",
        "type": "clarify",
        "context": "jane mentioned 'success' twice but never quantified it"
      }
    ],
    "speaker_count": 2,
    "speaker_names": {
      "speaker 1": "jane doe"
    },
    "transcript": [
      {
        "speaker": "you",
        "text": "thanks for making time today...",
        "timestamp": 0.42
      },
      {
        "speaker": "jane doe",
        "text": "happy to, let me share some context...",
        "timestamp": 3.88
      }
    ],
    "calendar_event_id": "abc123@google.com",
    "attendees": [
      { "email": "jane@acme.com", "name": "jane doe", "domain": "acme.com" }
    ]
  }
}

​
field notes

  • meddpicc, training, questions, attendees, and speaker_names are omitted when empty - do not assume these keys are always present
  • speaker_names is a map from the underlying diarization key to the resolved real name (e.g. "jane doe"). the key is whatever miniti uses internally to identify that speaker - it is not guaranteed to be a human-readable "speaker N" string, so treat it as an opaque id and look up names by key rather than parsing it. see speaker names for how miniti resolves them.
  • transcript[].speaker is the display label for that segment: the resolved real name when one is available, otherwise the display form of the diarization label, or "you" for your own mic. do not assume it matches the speaker_names key.
  • calendar_event_id is present only if the meeting was linked to a google calendar event
  • transcript[].timestamp is seconds from the start of the recording
  • date and end_time are ISO 8601 utc
  • keys are always json-sorted - downstream consumers can rely on stable ordering

​
receiver requirements

  • respond with any 2xx status - miniti logs non-2xx as failures but does not retry
  • respond within 10 seconds or miniti will time out
  • accept a payload up to ~a few mb for long meetings (transcript size dominates)

​
recipes

  • zapier / make / n8n - use their built-in “webhook by url” trigger and map fields from the meeting object
  • slack - forward summary + action items to a channel by transforming the payload in a small middle function
  • notion / airtable - create a row per meeting.saved event, mapping id, title, summary, action_items

​
see also