Outbound Phone Calling API

Overview

The Outbound Call API allows you to trigger an outbound call from your system to a specified phone number. The call can be customized with specific prompts, questions, and rules, and can include metadata about the customer. This API is useful for any situation where automated outbound calls are required.

Endpoint

POST https://phone.mylibby.ai/outbound

Request Payload

The request payload should be a JSON object containing the following fields:

{
  "name": "Chris",
  "introNotes": "We are an event planning company, Check and see if they are interested in event planning services",
  "toPhoneNumber": "+1xxxxxxxxxx",
  "fromPhoneNumber": "+1xxxxxxxxxx",
  "customQuestions": ["Are you interested in landscape photography for your event?", "Is there a time limit to the session you would like to impose?"],
  "customRules": ["Just be a friendly assistant", "Try to answer questions helpfully"],
  "formInfo": [
    {"Have you had a photographer before?":"Yes"}, 
    {"What's your budget?": "100 per month"}
  ]
}

Parameters

• name (string): The name of the person being called. This is used to personalize the conversation.
• toPhoneNumber (string): The phone number to call. This should be in E.164 format (e.g., "+1505XXXXXXX").
• fromPhoneNumber (string): The agent's phone number initiating the call. This should also be in E.164 format (e.g., "+1XXXXXXX").
• introNotes (string, optional): This is used to generate the introductory message or voicemail content. This will be used to start the conversation or leave a message if the recipient doesn't answer. It will not match the message, but rather be used in generating the message(s)
• introMessage (string, optional): Use this instead of introNotes if you want to define exactly how you want the assistant to saw after the user answers the phone.
• introVoicemail (string, optional): Use this instead of introNotes if you want to define exactly the voicemail you want the assistant to leave if the user does not answer the phone.
• customQuestions (array of strings, optional): A list of custom questions to ask during the call. If provided, these questions will override the existing set. If empty, the default questions will be used.
• customRules (array of strings, optional): Custom rules to guide the conversation. If provided, these will replace the default conversation rules.
• formInfo (array of objects): Metadata about the customer. This can include any relevant information that should be referenced during the call. Each object represents a key-value pair. Usually this information is taken in and dynamically set from a form.

Response

200

Call has been processed

{
    "message": "Outbound call processed"
}

Common Error Response

403 - Forbidden

Typically this indicates you did not add a token or the token is expired.

{
    "message": "Forbidden: Incorrect scopes"
}

404 - Not Found

Their is no agent for the outgoing number

{
    "message": "Agent not found for this number"
}

Authentication

Overview

The Cognito Token API is used to obtain an access token from Amazon Cognito by making a POST request to the OAuth2 token endpoint. This token can be used to authenticate requests to the Outbound API.

Endpoint

**POST https://libby.auth.us-east-1.amazoncognito.com/oauth2/token**

Request

Headers

• Content-Type: application/x-www-form-urlencoded

  Specifies that the request body is URL-encoded.

• Optional Cookie Cookie: XSRF-TOKEN={}

  The CSRF token for protecting against cross-site request forgery.

Body

The body of the request should be URL-encoded and include the following parameters:

• grant_type (string): The type of grant being requested. For client credentials, use client_credentials.
• client_id (string): The client ID for your application. This identifies your app to the authentication server.
• client_secret (string): The client secret associated with the client ID. This is used to authenticate your app.
• scope (string): The scope of the access being requested. This should be the resource server scope, e.g., external-api-resource-server-1/call.out.
  
  

const myHeaders = new Headers();
myHeaders.append("Content-Type", "application/x-www-form-urlencoded");
myHeaders.append("Cookie", "XSRF-TOKEN=0d12bcfe-f40f-4182-8c00-aa344553747f");

const urlencoded = new URLSearchParams();
urlencoded.append("grant_type", "client_credentials");
urlencoded.append("client_id", "your-client-id");
urlencoded.append("client_secret", "your-client-secret");
urlencoded.append("scope", "external-api-resource-server-1/call.out");

const requestOptions = {
  method: "POST",
  headers: myHeaders,
  body: urlencoded,
  redirect: "follow"
};

fetch("https://libby.auth.us-east-1.amazoncognito.com/oauth2/token", requestOptions)
  .then((response) => response.text())
  .then((result) => console.log(result))
  .catch((error) => console.error(error));