Read Avoma meetings, recordings, transcripts, AI notes, insights, scorecards and users, and submit call recordings for analysis.
Avoma ships in the w6w first-party pack. It declares 19 actions, 3 health checks, and the host runs its code in a sandbox that never sees the credential.
io.w6w.avomaAvoma is an AI meeting assistant that records, transcribes and summarizes customer conversations. This app lets a workflow pull that conversation data out: list meetings in a date window by attendee, CRM association or call type, then read each meeting’s transcript, AI-generated notes, keyword insights, sentiment, topic segments and expiring recording URLs.
It also covers the coaching side — scorecard templates and the evaluations completed against meetings — plus the organization’s users and smart categories. For teams with their own dialer, a workflow can submit a call recording to Avoma for transcription and analysis, look the call up again by its external id, and remove Avoma’s notetaker bot from a meeting it is attending.
Avoma’s API requires a date window on its main list endpoints and rate-limits to 60 requests a minute, so the actions validate dates up front and page by the vendor’s own next-page URL. Avoma publishes no status page, so health is tracked by an unauthenticated reachability check on the API plus the credential check.
Three routes to the same 19 actions. The Workflow tab is generated from Avoma's own manifest and carries its real ids, so it is copy-pasteable; the Code and CLI examples are the same call for any action on any app, so every app-specific value in them is a blank you fill in.
call-create Submit a call recording from your own dialer to Avoma for transcription and analysis.
call-list List calls (from a connected dialer or created through the API) that started inside a date window.
meeting-get Fetch one meeting: subject, times, attendees, outcome, and which of the recording, transcript and notes are ready.
meeting-insights-get AI-generated notes, keyword occurrences and speakers for a completed meeting.
meeting-list List meetings that started inside a date window, with optional attendee, CRM and call/video filters.
meeting-segment-list Time ranges (seconds) of the agenda, intro, demo, pricing, objection, pain point and next-steps segments of a meeting.
meeting-sentiment-list Overall sentiment and per-time-range sentiment scores for a meeting.
note-list List AI-generated meeting notes in a date window, optionally for one meeting or one smart category, as JSON, HTML or Markdown.
recording-get Get expiring audio and video download URLs for a meeting's recording. If it is still processing, the URLs are absent and a message says so.
scorecard-evaluation-list List scorecard evaluations (scores, answers and justifications) for meetings, filtered by date, meeting, template or the person scored.
smart-category-list List the organization's smart categories (keywords, prompts and AI-note settings used to tag and extract notes).
transcription-get Fetch a transcript by its transcription UUID: speakers, timestamped paragraphs and the WebVTT URL.
transcription-list Fetch the transcript for one meeting, or list transcripts for meetings in a date window. Each has speakers and timestamped paragraphs.
A workflow step names the app and the action, and the editor fills in the
connection when you pick one. This is the Step shape from the
workflow spec, carrying Avoma's real ids.
{
"manifestVersion": "2",
"name": "avoma-example",
"steps": [
{
"id": "call-create",
"uses": {
"app": "io.w6w.avoma",
"action": "call-create",
"connection": "conn_YOUR_CONNECTION_ID"
},
"with": {
"externalId": "<externalId>",
"userEmail": "<userEmail>",
"frm": "<frm>"
}
}
]
}call-create call-get call-list meeting-drop meeting-get +14 more actions available
Every app-specific value here is a blank you have to fill in. An
app action is reached through the connection that authenticates it, so the
address is a connection id, not the app id — and connections belong to your account,
so a public page cannot know yours. Create one for Avoma, then fill in
the three blanks: conn_YOUR_CONNECTION_ID, the action key, and the
parameters that action declares. The call itself is real — the shape is transcribed
from the studio's own snippet builder, which prints the same kind of blanks — but
nothing in it is specific to Avoma. The Workflow tab is where this app's
real ids are.
npm install @w6w/sdkyarn add @w6w/sdkpnpm add @w6w/sdkdeno add npm:@w6w/sdkimport { W6wClient, isActionRun } from "@w6w/sdk";
// Reads W6W_BASE_URL and W6W_TOKEN from the environment when omitted.
const client = new W6wClient();
const envelope = await client.run({
urn: "conn_YOUR_CONNECTION_ID",
action: "call-create",
payload: {
externalId: "<value>",
userEmail: "<value>",
frm: "<value>",
to: "<value>",
startAt: "<value>",
// endAt: "<value>",
recordingUrl: "<value>",
direction: "<value>",
source: "<value>",
participants: "<value>",
// frmName: "<value>",
// toName: "<value>",
// answered: "<value>",
// isVoicemail: "<value>",
},
});
if (isActionRun(envelope)) console.log(envelope.value); npm install -g @w6w/cli w6w run conn_YOUR_CONNECTION_ID --action call-create --payload '{"externalId":"<value>","userEmail":"<value>","frm":"<value>","to":"<value>","startAt":"<value>","recordingUrl":"<value>","direction":"<value>","source":"<value>","participants":"<value>"}' Give an AI agent Avoma — without giving it Avoma's credentials. One MCP endpoint exposes every app, function and workflow the caller is entitled to, as tools it can discover and run. Access is granted per team while we onboard.
One tool call{
"name": "w6w_invoke",
"arguments": {
"ref": "app:io.w6w.avoma#call-create",
"input": {
"externalId": "<externalId>",
"userEmail": "<userEmail>",
"frm": "<frm>"
}
}
}
Every tool names its target with a single ref. The
app: form above doesn't name a connection at all — the
host resolves which of the caller's Avoma connections to sign
with, and refuses rather than guesses when the answer is ambiguous.
The token is attached host-side, at the moment of the call. It is never a tool argument, never in the model's context, and never in a transcript — so a prompt injection has nothing to exfiltrate.
Tools are derived per end user from what that person has actually connected and is entitled to — not one shared bot identity carrying the union of everyone's access.
Multi-step work runs on the workflow engine and returns a run handle the agent can poll — retries, branching and state survive the conversation that started them.
Avoma's declared health checks are on the surface too, so an agent can tell "the vendor is down" from "your credential expired" before it burns a retry on either.
The MCP surface is part of the hosted platform. Avoma itself is MIT, and the runtime that executes it is source-available (FSL).
Avoma declares its own checks, so its health is a property of the app rather than something the host guesses at.
Unauthenticated GET https://api.avoma.com/v1/users/. A 401 carrying Avoma's JSON `{"detail": …}` passes — it proves the API and its auth layer are answering. Credential validity is the derived `auth:*` check's job.