First-party app
Retell AI Retell AI

Retell AI

Place outbound phone and web calls, batch-dial numbers, and search call history with Retell AI's voice-agent platform.

stable AI & Machine LearningCommunication

About

Retell AI ships in the w6w first-party pack. It declares 9 actions, 2 health checks, and the host runs its code in a sandbox that never sees the credential.

App id
io.w6w.retellai
Version
0.1.1
Author
w6w
Licence
MIT
Categories
AI & Machine Learning · Communication

Overview

Retell AI runs AI voice and chat agents over phone and browser calls, and this app is what places and manages the calls an already-configured agent makes: dialing an outbound phone call, starting a browser-based call session, and queuing a batch of calls against a list of numbers with an optional calling window. It is built for workflows that trigger a call in response to something else happening — a new lead, a scheduled reminder — using an agent already set up in the Retell dashboard, rather than for configuring the agent’s own voice, prompt, or tools.

Call results can be searched and read back, including the transcript, recording, and analysis once a call finishes — calls run and finish asynchronously, so reading one back is a matter of polling until it completes. Available phone numbers and voices can be listed for use when placing a call, and current call concurrency can be checked directly, which is the real ceiling on how many calls can run at once.

Build with Retell AI

Three routes to the same 9 actions. The Workflow tab is generated from Retell AI'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.

Create Batch Call

create-batch-call

Queue a batch of outbound phone calls from one number, immediately or scheduled.

Create Phone Call

create-phone-call

Start an outbound phone call from a number this account owns or has imported.

Create Web Call

create-web-call

Start a browser-based call session with an agent, returning a client access token.

Get Call

get-call

Retrieve a call's current status, transcript and recording (once it has ended).

Get Concurrency

get-concurrency

Read the org's current concurrent-call usage and limit.

List Agents

list-agents

List voice or chat agents in the workspace, paginated.

List Calls

list-calls

Search calls by agent, status or number, newest first by default.

List Phone Numbers

list-phone-numbers

List phone numbers owned by or imported into this account.

List Voices

list-voices

List every voice available to this account. Not paginated — Retell returns the whole catalog in one call.

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 Retell AI's real ids.

{
  "manifestVersion": "2",
  "name": "retellai-example",
  "steps": [
    {
      "id": "create-batch-call",
      "uses": {
        "app": "io.w6w.retellai",
        "action": "create-batch-call",
        "connection": "conn_YOUR_CONNECTION_ID"
      },
      "with": {
        "fromNumber": "<fromNumber>",
        "tasks": "<tasks>"
      }
    }
  ]
}

Here are some of the things you can do

  • Create Batch Call

    perform
    create-batch-call
  • Create Phone Call

    perform
    create-phone-call
  • Create Web Call

    perform
    create-web-call
  • Get Call

    read
    get-call
  • Get Concurrency

    read
    get-concurrency

+4 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 Retell AI, 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 Retell AI. The Workflow tab is where this app's real ids are.

Install
npm install @w6w/sdk
yarn add @w6w/sdk
pnpm add @w6w/sdk
deno add npm:@w6w/sdk
Code
import { 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: "create-batch-call",
  payload: {
    fromNumber: "<value>",
    tasks: "<value>",
    // name: "<value>",
    // triggerTimestamp: "<value>",
    // reservedConcurrency: "<value>",
    // timeWindows: "<value>",
    // timeWindowTimezone: "<value>",
  },
});

if (isActionRun(envelope)) console.log(envelope.value);
Install the CLI
npm install -g @w6w/cli
CLI
w6w run conn_YOUR_CONNECTION_ID --action create-batch-call --payload '{"fromNumber":"<value>","tasks":"<value>"}'

Give an AI agent Retell AI — without giving it Retell AI'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.retellai#create-batch-call",
    "input": {
      "fromNumber": "<fromNumber>",
      "tasks": "<tasks>"
    }
  }
}

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 Retell AI connections to sign with, and refuses rather than guesses when the answer is ambiguous.

What the agent gets

Credentials it can't read

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.

A tool surface scoped to the caller

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.

A durable workflow in one call

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.

Health-aware discovery

Retell AI'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. Retell AI itself is MIT, and the runtime that executes it is source-available (FSL).

Request MCP access

Health checks

Retell AI declares its own checks, so its health is a property of the app rather than something the host guesses at.

service

Retell AI platform status

Component status from status.retellai.com, including the API, Web Call, Phone Call and Batch Call components this app's actions actually use.

quota

Concurrent-call headroom

Current vs. maximum concurrent calls, from GET /get-concurrency.