# Welcome to iX Hello

**iX Hello™ is part of the @Concentrix Intelligent Experience (iX) Technology suite. Whether you are a developer or business user; whether you are an experienced user or just starting your journey with iX Hello, our documentation provides you with the comprehensive information and knowledge needed to build exciting solutions for your customers’ needs.**

Discover the power and versatility of iX Hello through its two main modules: 

* Robust enterprise conversational assistants for customers
* Rapid Gen AI virtual assistants for employees

Each module is engineered to enhance your conversational AI capabilities with ease and efficiency. Click the links below to dive deeper into each module's documentation:

<table data-card-size="large" data-view="cards"><thead><tr><th></th><th></th><th></th><th data-hidden data-card-cover data-type="files"></th></tr></thead><tbody><tr><td><strong>Customers</strong></td><td><p></p><ul><li>Supports multi-modal applications deployable in minutes.</li><li>Integration with common enterprise applications like CRMs, ERPs, and ITSMs.</li><li>Template-driven quick start for common enterprise use cases.</li><li>Blueprints for industry use cases</li></ul></td><td></td><td></td></tr><tr><td><strong>Employees</strong></td><td><p></p><ul><li>Rapid Gen AI deployments and easy adoption.</li><li>Out-of-the-box integrations with existing systems.</li><li>Customizable chat interface tailored to your brand.</li><li>Enhanced privacy and data security measures.</li></ul></td><td></td><td></td></tr></tbody></table>

{% content-ref url="/spaces/-M8XHvUsfyTUFLvToHqD" %}
[iX Hello Customer](https://docs.ixhello.com/ixhc/)
{% endcontent-ref %}

{% content-ref url="/spaces/l2dmTEKSqnEOjqqng3nd" %}
[iX Hello Employee](https://docs.ixhello.com/ixhello/)
{% endcontent-ref %}

{% content-ref url="/spaces/Fe4cpY1jy0FJ4kgC7wSx" %}
[iX Hello Customer v2](https://docs.ixhello.com/ixhc2/)
{% endcontent-ref %}


# Welcome to ix Hello (original)

## Welcome to the official ix Hello Documentation.

**ix Hello™ is part of the @Concentrix Intelligent Experience (iX) Technology suite. Whether you are a developer or a business user; whether you are an experienced ix Hello user or just starting your journey with ix Hello - our documentation provides you with the comprehensive information and knowledge needed to build exciting solutions for your customers’ needs.**

Discover the power and versatility of ix Hello through its two main modules for creating robust enterprise conversational assistants for customers and rapid Gen AI virtual assistants for employees. Each module is engineered to enhance your conversational AI capabilities with ease and efficiency. Click the links below to dive deeper into each module's documentation.

## **Robust enterprise conversational assistants for customers**

| <ul><li>Supports multi-modal applications deployable in minutes.</li><li>Integration with common enterprise applications like CRMs, ERPs, and ITSMs.</li><li>Template-driven quick start for common enterprise use cases.</li><li>Blueprints for industry use cases</li></ul> |
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

Click the link below to discover more:

{% content-ref url="/spaces/-M8XHvUsfyTUFLvToHqD" %}
[iX Hello Customer](https://docs.ixhello.com/ixhc/)
{% endcontent-ref %}

## &#x20;**Rapid Gen AI virtual assistants for employees**

| <ul><li>Rapid Gen AI deployments and easy adoption.</li><li>Out-of-the-box integrations with existing systems.</li><li>Customizable chat interface tailored to your brand.</li><li>Enhanced privacy and data security measures.</li></ul> |
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

Click the link below to discover more:

{% content-ref url="/spaces/l2dmTEKSqnEOjqqng3nd" %}
[iX Hello Employee](https://docs.ixhello.com/ixhello/)
{% endcontent-ref %}


# Videos

The following videos offer quick entries into the world of ix Hello


# EU Instance Rollout for iX Hello

An EU instance for iX Hello is now live in Sweden, which includes both employee-facing and customer-facing AI Assistants.

To comply with GDPR and better serve EU-based users, Concentrix has launched a dedicated **EU instance** of iX Hello. This rollout provides enhanced data privacy, faster performance, and legal compliance for organizations operating in the European Union.&#x20;

## **What’s New**&#x20;

### **EU-Based Hosting**&#x20;

A separate environment is now live in **Sweden**, which includes both:&#x20;

* **Employees-facing**: [https://eu.employee.ixhello.com](https://eu.employee.ixhello.com/)&#x20;
* **Customers-facing**: [https://eu.customer.ixhello.com](https://eu.customer.ixhello.com/)&#x20;

### **Centralized Login**&#x20;

Login continues through a global Azure B2C system (US-based), which determines your organization’s routing after successful authentication.&#x20;

### **Azure OpenAI in the EU**&#x20;

All AI-powered features (e.g., chat assistants) are now supported via **Azure OpenAI deployed in Sweden**, ensuring regional compliance.&#x20;

## **Why This Matters**&#x20;

* **GDPR Compliance:** Keeps EU user data within EU borders.&#x20;
* **Improved Performance:** Reduced latency for EU-based access.&#x20;
* **Trust & Compliance:** Aligns with enterprise expectations for data sovereignty.&#x20;
* **No Cross-Region Access:** EU and US instances are separate; AI Assistants and data do not sync across regions.&#x20;

## **Useful Tips for End-Users**&#x20;

### **How to Know Which URL to Use**&#x20;

If your organization is based in the EU, please bookmark and use the following:&#x20;

* **Employee-facing:** [https://eu.employee.ixhello.com](https://eu.employee.ixhello.com/)&#x20;
* **Customer-facing:** [https://eu.customer.ixhello.com](https://eu.customer.ixhello.com/)&#x20;

If you're unsure which region you're assigned to, reach out to your team lead or support.&#x20;

### **What to Expect During Login**&#x20;

You might notice a brief redirect to a US-hosted login page — this is expected. The global login system identifies your organization and routes you to the right environment.&#x20;

User experience and data remain fully in the **EU region** if your organization is EU-based.&#x20;

### **Traveling? Don’t Worry**&#x20;

Even if you're accessing iX Hello from outside the EU (e.g., during travel), you’ll still be routed to the EU environment, based on your organizational profile.&#x20;

## **Known Limitations**&#x20;

* Assistants and content created in the US environment are **NOT** available in the EU.&#x20;
* The current EU rollout is **English-only**. Multi-language support is planned soon.&#x20;
* Some advanced features may launch in the EU region later than in the US.&#x20;

## **Troubleshooting & Support**&#x20;

* **Routing Issues:** Check that your organization’s residency is set correctly.&#x20;
* **Login Problems:** Retry using your full email address. Confirm that you're using the correct URL.&#x20;
* **Need Help?** Email us at <ix_hello_support@concentrix.com>


# Glossary

* **Account Setup**: Holds user profile: name, contact information, password. **M**ulti **F**actor **A**uthentication is turned on here.
* **Agent**: (1) A human tasked to help a customer inquiry \
  (2) AI bot designed to help the user with any and all tasks, enterprise or otherwise related.
* **APL**: **A**lexa **P**resentation **L**anguage. Software to used when an App runs on Alexa device.
* **App**: An AI or text/speech recognition-based system that provides a service to a user.
* **App Blueprints**: A pre-built template for an App that contains intents, utterances, methods, etc. Blueprints often only need connection information to be fully functional.
* **App Template**: A reusable framework for a custom App. Templates are JSON files that can be created, shared, modified, and converted into a custom app.&#x20;
* **AI**: Computer systems capable of performing complex tasks such as reasoning, making decisions, or solving problems.
* **AI Content**: The backend that houses and manages all text training data and RAG-trained content for a specific app/assistant
* **Assistant**: An AI or text/speech recognition-based system that provides a service to a user.
* **Basic Mode**: Simplified bot creation page for employees
* **Blueprints**: See App Blueprints above.
* **Bot**: AI system designed to help the user with any and all tasks, enterprise or otherwise related
* **Category**: A group of assistants.
* **Channel**: The method through which end users experience, interact, or engage with an app or assistant
* **Chatbot**: An AI-based service that simulates human conversation.
* **Chunk Size**: The maximum number of characters that a chunk can contain.
* **Chunk overlap:** The number of characters that should overlap between two adjacent chunks.
* **Collections**: List of Blueprints
* **Connection Admin**: Page to set Log in configurations for external software such as Amazon, Office 365, etc.
* **Content**: All required training and structural data associated with a particular app/assistant is stored, categorized and accessed for use.
* **Conversations**: Records of user interactions with an App
* **Data Sources**: Web sites, data bases, etc. with will hold information to be used by the bot.
* **Description**: Text used to personalize a bot's responses.
* **Digital Human**: A life-like avatar representing an AI assistant
* **Entity**: Entities are specific pieces of information extracted from user input that represent real-world objects, concepts, or values. For example, if a caller in a flight reservation App states "I want to fly to London", the Entity would be "London"
* **Input**:  Inputs enable channels (e.g. Alexa or Lex) to pass in Slots (parameters) associated with an Intent that can be pre-processed and mapped to one or more internal method parameters. Inputs can contain multiple Input parameters
* **Instruction**: Language used to describe tasks to an assistant.
* **Integration**: Makes a connection between an iX Hello bot and an external data source, cloud service, web page etc. Account passwords or keys are often entered here.&#x20;
* **Intent:** A user's reasons for interacting with an App. For example, booking a flight, getting weather, etc. Intents should be given a descriptive name, often a verb and a noun, such as “BookFlight".
* **Knowledgebase**: A repository for uploaded documents or images for an assistant to scan or use
* **Large Language Models:** Systems designed to understand and generate human language.
* **Methods**: Instructions on how the bot will treat information retrieved in the search.
* **MFA**: (**M**ulti **F**actor **A**uthentication) Security measure that requires that the user submit a code from either their phone or email to continue their task. Turned on under **Account Setup** > **Profile**.
* **Public Apps**: Shared Apps created by iX Hello users.
* **Publish**: Publishing creates the working app.&#x20;
* **Q\&A**: **Q**uestion and **A**nswer
* **RAG bot**: **R**etrieval **A**nd **G**eneration bots are AI powered chatbots that provide information based on a pre-defined source such as a web page.
* **Session Attributes**: Parameters added to a method to assist in processing.
* **Slot**: Slots are place holders for a required component of an utterance. For instance, if a user's intent is to book a flight, they would need to give a destination. The utterance would be written as "I want to fly to \<flight destination>" with \<flight destination> standing in for any number of possible destinations.
* **Slot Type**: Slot types are user defined representations of slots of type custom. Slot Types can be assigned to multiple custom slots and they have a collection of Slot Values and Synonyms. Slot types are given a name, such as "Destination" and have related Slot Values like "Chicago" or "Dallas" and can be assigned to custom slots.
* **Studio Mode**: Site to build custom AI bots&#x20;
* **Survey**: Bot designed to retrieve information from a user. A Survey is AI-led rather than user-led.
* **Template**: See App Template.&#x20;
* **Tool**: An entity that an assistant uses to help complete a task, such as a calculator.
* **Training Contents**: Web sites, text files, etc. supplied to an App to assist with responses.
* **Utterance**: Utterances are what you expect the user to say to your Custom App. For example: "Book me a flight" or "I need to make an appointment".
* **Web Chat**: A web page feature that allows a user to ask questions and get responses from a human or AI assistant.


# About iX Hello

iX Hello offers an easy No-Code platform for building powerful, dynamic, and secure enterprise conversational assistants such as chatbots or intelligent personal assistants. A user can create a voice assistant or App and securely connect it to an external enterprise data source or internal content source. The voice assistant or App can then be launched to your desired platforms including Amazon Alexa, Amazon LEX, Google Assistant, or Twilio SMS.

<figure><img src="/files/mtcZTcImdGn356kuTj0z" alt=""><figcaption></figcaption></figure>

The user can also select from one of our pre-built templates of common enterprise use cases to jump-start creation of their voice assistant.

{% tabs %}
{% tab title="Overview" %}

* It is a no-code Conversational AI platform.
* The user experience is fast, simple, and easy to use by non-developers.
* It allows IT Admins to create modern multi-modal applications in minutes and deploy to enterprise users or consumers.
* It is cost effective and helps eliminate expensive data scientists, conversational AI developers, or weeks of programming.
* Supports most common CAI channels - Amazon Alexa (100 million devises), Google Assistant (1 billion devices\*), Skype, Slack, and SMS.
* Provides out-of-the-box integration to most common Enterprise Applications like CRMs (e.g., Salesforce), ERPs (e.g., SAP), ITSMs (e.g., ServiceNow) platforms.
* Provides a configurable REST API interface that enables secure connections to most enterprise systems.
  {% endtab %}

{% tab title="Benefits" %}

* Rapid, no-code voice-enablement of enterprise applications.
* Operates on user's Network/Private Cloud.
* Easy to integrate new “Connectors” to new systems via REST API.
* Helps users by standardizing voice integration platform.
* Provides API for developers to integrate into.
* Shares tested and secure architecture with users.
  {% endtab %}
  {% endtabs %}

**iX Hello Components**

<table data-card-size="large" data-view="cards"><thead><tr><th></th><th></th><th></th><th></th></tr></thead><tbody><tr><td><strong>Assistants</strong></td><td>Create and manage digital assistants to power voice and conversational experiences across various channels like Alexa.</td><td></td><td></td></tr><tr><td><strong>Connections</strong></td><td>Establish and manage secure connections to enterprise systems like Salesforce or custom systems via API for use in your Assistants.</td><td></td><td></td></tr><tr><td><strong>Content Management</strong></td><td>Create and manage various types of enterprise contents like FAQs, Surveys, and Videos for use in your digital assistants.</td><td></td><td></td></tr><tr><td><strong>Assistant Templates</strong></td><td>Start with pre-built templates for your Digital Assistant. This saves time to easily implement common use cases.</td><td></td><td></td></tr></tbody></table>


# About iX Hello

iX Hello offers an easy No-Code platform for building powerful, dynamic, and secure enterprise conversational assistants such as chatbots or intelligent personal assistants. A user can create a voice assistant or App and securely connect it to an external enterprise data source or internal content source. The voice assistant or App can then be launched to your desired platforms including Amazon Alexa, Amazon LEX, Google Assistant or Twilio SMS.

<figure><img src="/files/mtcZTcImdGn356kuTj0z" alt=""><figcaption></figcaption></figure>

The user can also select from one of our pre-built templates of common enterprise use cases to jump-start creation of your voice assistant.

{% tabs %}
{% tab title="Overview" %}

* It is a no-code Conversational AI platform.
* The user experience is fast, simple and easy to use by non-developers.
* It allows IT Admins to create modern multi-modal applications in minutes and deploy to enterprise users or consumers.
* It is cost effective, and it helps to eliminate expensive data scientists, conversational AI developers or weeks of programming.
* Supports most common CAI channels - Amazon Alexa (100 million devises), Google Assistant (1 billion devices\*), Skype, Slack, and SMS.
* Provides out-of-the-box integration to most common Enterprise Applications like CRMs (e.g., Salesforce), ERPs (e.g., SAP), ITSMs (e.g., ServiceNow) platforms.
* Provides a configurable REST API interface that enables secure connections to most enterprise systems.
* Offers robust **Multilanguage support**, enabling users to interact with the platform in their preferred language.
  {% endtab %}

{% tab title="Benefits" %}

* Rapid, no-code voice-enablement of enterprise applications.
* Operates on users Network/Private Cloud.
* Easy to integrate new “Connectors” to new systems via REST API.
* Helps users by standardizing voice integration platform.
* Provides API for developers to integrate into
* Shares tested and secure architecture with users.
* Cater to diverse user bases across different regions.
* Allows users to engage in their preferred language.
  {% endtab %}
  {% endtabs %}

**iX Hello Components**

<table data-card-size="large" data-view="cards"><thead><tr><th></th><th></th><th></th><th></th></tr></thead><tbody><tr><td><strong>Assistants</strong></td><td>Create and manage digital assistants to power voice and conversational experiences across various channels like Alexa.</td><td></td><td></td></tr><tr><td><strong>Connections</strong></td><td>Establish and manage secure connections to enterprise systems like Salesforce or custom systems via API for use in your Assistants.</td><td></td><td></td></tr><tr><td><strong>Content Management</strong></td><td>Create and managed various types of enterprise contents like FAQ, Surveys, and Video for use in your digital assistants.</td><td></td><td></td></tr><tr><td><strong>Assistant Templates</strong></td><td>Started with pre-built templates for your Digital Assistant. This saves time to easily implement common use cases.</td><td></td><td></td></tr></tbody></table>


# Multilanguage Support - iX Hello

**Multilanguage support** refers to a system or software feature that allows content, interfaces, and interactions to be presented in multiple languages. It enables users from different linguistic backgrounds to use the product or service in their **native or preferred language**, improving usability and accessibility.

#### Key Aspects of Multilanguage Support:

1. **User Interface Translation**\
   Menus, buttons, labels, and system messages are translated into various languages.
2. **Content Localization**\
   Text, images, and other media are adapted to fit cultural and language norms of different regions.
3. **Input & Output Handling**\
   Users can input text in one language and receive responses or outputs in the same or another selected language.
4. **Language Switching**\
   Users can change the display language easily, often through a dropdown or settings menu.
5. **Languages supported**                                                                                                                                          English, LATAM Spanish, Korean, French, German, EU Spanish, Br Portuguese, Turkish, Simplified Chinese & Japanese, Malay & Vietnamese, Hindi, Arabic and Italian.
6. **Backend Flexibility**\
   The system is built to support language files or databases for multiple languages without requiring major code changes.

**Why It Matters:**

* **Inclusivity**: Reaches a broader, global audience.
* **Accessibility**: Helps non-native speakers use the platform comfortably.
* **Customer Experience**: Increases user satisfaction and engagement.
* **Compliance**: Meets regional language laws or business requirements.

**How to use Multilangauge in iX Hello**

**Pre-requisite:**

#### Creating an iX Hello Account

To get started with iX Hello, it's essential to have an iX Hello account. Follow these steps to set up your account:

1. **Account Creation**:
   * Contact our support team to initiate your account creation process.
   * Our team will assist you in creating an account.
2. **Sign Up Process**:
   * Access the sign-up page by navigating to [iX Hello Sign-Up](https://bots.ixhello.com/).
   * Click on "***Sign up now***".
3. **Choosing a Preferred Language**:
   * On the Login page, locate the language drop-down menu at the bottom right corner.
   * This menu allows you to select from a list of available languages.
   * **Hint**: After selecting a language, the interface automatically updates to your chosen language, providing a seamless user experience.

<figure><img src="/files/662XQCdHDX6OF29Ybqo8" alt=""><figcaption></figcaption></figure>

1. **Navigation Consistency**:
   * The language drop-down menu is persistently located at the bottom right corner across all menus and screens within the platform.

Utilize these steps to navigate effectively and customize your experience in iX Hello.

<figure><img src="/files/M9UXvxraqQIO6OV9rrUk" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/gT6ZAmJhCbqFbDDoDO2N" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/eNlGAPgde0i5GHDNHzFc" alt=""><figcaption></figcaption></figure>


# About ix Hello - original

A user can create a voice assistant or App, securely connect it to an external enterprise data source or internal content source. The voice assistant or App can then be launched to your desired platforms including Amazon Alexa, Amazon LEX, Google Assistant or Twilio SMS.

**ix Hello Overview**

* It is a no-code Conversational AI platform.
* The user experience is fast, simple and easy to use by non-developers.
* It allows IT Admins to create modern multi-modal applications in minutes and deploy to enterprise users or consumers.
* It is cost effective, and it helps to eliminate expensive data scientists, conversational AI developers or weeks of programming.
* Supports most common CAI channels - Amazon Alexa (100 million devises), Google Assistant (1 billion devices\*), Skype, Slack, and SMS.
* "Provides out-of-the-box integration to most common Enterprise Applications like CRMs (e.g., Salesforce) and ITSMs (e.g., ServiceNow) platforms.
* Provides a configurable REST API interface that enables secure connections to most enterprise systems.

‌ix Hello offers an easy No-Code platform for building powerful, dynamic, and secure enterprise conversational assistants such as chatbots or intelligent personal assistants.

The user can also select from one of our pre-built templates of common enterprise use cases to jump-start creation of your voice assistant. Select an option below to begin creating your voice assistant.

**ix Hello Platform Benefits**

* Rapid, no-code voice-enablement of enterprise applications.
* Operates on users Network/Private Cloud.
* Easy to integrate new “Connectors” to new systems via REST API.
* Helps users by standardizing voice integration platform.
* Provides API for developers to integrate into
* Shares tested and secure architecture with users.

**ix Hello Components**

**Assistants**

‌Create and manage digital assistants to power voice and conversational experiences across various channels like Alexa.

**Connections**

Establish and manage secure connections to enterprise systems like Salesforce or custom systems via API for use in your Assistants.

‌**Content Management**

‌Create and managed various types of enterprise contents like FAQ, Surveys, and Video for use in your digital assistants.

‌**Assistant Templates**

‌Get started with a pre-built template for your Digital Assistant. This saves time to easily implement common use cases.


# Videos

The following videos will assist you in learning how to use ix Hello.

### How to use Blueprints

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-M8XHvUsfyTUFLvToHqD%2Fuploads%2FTKpd35PvOGoUH7InbcG6%2FixHello%20-%20How%20to%20Use%20BlueprintsV2_FINAL.mp4?alt=media&token=1fce2699-4531-45e3-b3ea-82807fb30445>" %}

### Creating a Benefits Assistant

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-M8XHvUsfyTUFLvToHqD%2Fuploads%2FBwmQLVPRqiclaCVG96HD%2FCB%20How%20to%20Create%20a%20Benefits%20Assistant%20with%20Blueprints_FINAL.mp4?alt=media&token=ca90acad-7536-46e0-b414-dda3f4954b1b>" %}

### Creating an Appointment Booking Assistant

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-M8XHvUsfyTUFLvToHqD%2Fuploads%2FaHfDmdLqZis0BMEtgATf%2FCB%20How%20to%20Create%20an%20Appointment%20Booking%20%20%20Assistant%20with%20Blueprints_FINAL.mp4?alt=media&token=3ce64c8c-2942-4d98-8e8c-c6d2dbeb4b58>" %}

### Creating a Reminder Assistant

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-M8XHvUsfyTUFLvToHqD%2Fuploads%2FE7dAidHsNvEb7MWEZOpl%2FHow%20to%20Create%20a%20Reminder%20Assistant%20with%20Blueprints_FINAL.mp4?alt=media&token=375e3b5b-ec93-417f-9aa8-ea6939623209>" %}

### Creating a Snow Day Assistant

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-M8XHvUsfyTUFLvToHqD%2Fuploads%2FlciwIezYcddtzSejLOuY%2FHow%20to%20Create%20a%20Snow%20Days%20Assistant%20with%20Blueprints_FINAL%20(1).mp4?alt=media&token=038061f3-7517-4f74-8f1b-29ea14bfe624>" %}

### Interacting with web chat, phone, and SMS

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-M8XHvUsfyTUFLvToHqD%2Fuploads%2FUe8w29tBhnp54H6vt72H%2FCB%20How%20toCreate%20and%20Activate%20Assistant%20to%20Interact%20with%20webchat%2C%20phone%20call%20and%20SMS_FINAL.mp4?alt=media&token=92163393-68b4-49f9-90fc-644bb0efc415>" %}


# iX Hello: Create Account - original

## Create an Account in iX Hello

To create an account in iX Hello, browse to the “<https://bots.ixhello.com/>” URL and click “***Sign up now***”.

<figure><img src="/files/J9FNlCqm8TN28mymacXD" alt="" width="375"><figcaption></figcaption></figure>

You are directed to a “***Sign up now***”. On the Register page, fill in all the form fields and “***Login***”.

<figure><img src="/files/2qY0nrsDr8OCeG6IQJg4" alt="" width="184"><figcaption></figcaption></figure>

Your Email ID will be verified when you click on Send Verification Code.

<figure><img src="/files/bD98MU3jUH9okziZX5KM" alt="" width="375"><figcaption></figcaption></figure>

An email with the verification code will be sent to your registered email id as below:

It may land in your junk or spam folders so be sure to check those folders in addition to your inbox.

<figure><img src="/files/23MtWoA4UgRE7X2SbIT9" alt="" width="375"><figcaption></figcaption></figure>

Enter the verification code and characters as below example and click verify code

<figure><img src="/files/w8ysbRq1lfwtfDYbRztC" alt="" width="293"><figcaption></figcaption></figure>

Once clicked on Verify code, you will see a notification, that your email ID is verified.

After Email ID is verified, kindly fill in the necessary details and click "Create", your iX Hello account will be created and click continue.

<figure><img src="/files/hRoVWoG5YoV1xuArH6Ln" alt="" width="296"><figcaption></figcaption></figure>

<figure><img src="/files/nDSzWYM91oAbtm9I1RSy" alt="" width="296"><figcaption><p>, </p></figcaption></figure>

Once you click continue you will be redirected to the login page, kindly use your email id and newly created password to login.

<figure><img src="/files/I59ewciIwV8NNYZ6g9Qx" alt="" width="563"><figcaption></figcaption></figure>

After creating the account and logging in, you will land on the iX Hello screen as shown below:

<figure><img src="/files/R09qqH9eOlcQIRq3Dgpv" alt=""><figcaption></figcaption></figure>

After registration, an OTP will be sent to the registered email address. Please check your email for the OTP and paste it to confirm your email. Note: It may land in your junk/spam folder.

<figure><img src="/files/4f51yXBwC8zS3GzXXO1w" alt="" width="374"><figcaption></figcaption></figure>

After login process, click on send verification code, a code will be sent to the registered email address. Please check your email for the code and paste it to confirm your email. Note: It may land in your junk/spam folder.

<figure><img src="/files/yVXRrIH1KB0I8yFcMzDP" alt="" width="371"><figcaption></figcaption></figure>

<figure><img src="/files/mYDrNdynug09OAwhYuEv" alt="" width="375"><figcaption></figcaption></figure>

After confirming the OTP, it verifies your email and will open a new tab <https://bots.ixhello.com/>


# iX Hello: Create Account

{% embed url="<https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-M8XHvUsfyTUFLvToHqD%2Fuploads%2Ft7aWOyRgpyMiIWiWY97W%2FHow%20to%20Create%20an%20Account%20in%20ixHello%20and%20Join%20an%20Existing%20Organization.mp4?alt=media&token=df73cace-4468-4bec-99ed-9f164fba47fb>" %}

## Create an Account in iX Hello

1. To create an account in iX Hello, browse to the “<https://bots.ixhello.com/>” URL and click “***Sign up now***”.

<figure><img src="/files/J9FNlCqm8TN28mymacXD" alt="" width="375"><figcaption></figcaption></figure>

2. You are directed to a “***Sign up now***” page. Fill in all the form fields.

<figure><img src="/files/2qY0nrsDr8OCeG6IQJg4" alt="" width="184"><figcaption></figcaption></figure>

3. Press "**Send Verification Code**". This will verify your email address.
4. An email with the verification code will be sent to your registered email id. It may land in your junk or spam folders so be sure to check those folders in addition to your inbox.

<figure><img src="/files/23MtWoA4UgRE7X2SbIT9" alt="" width="375"><figcaption></figcaption></figure>

5. Enter the verification code and characters as below example and click verify code

<figure><img src="/files/w8ysbRq1lfwtfDYbRztC" alt="" width="293"><figcaption></figcaption></figure>

6. Once clicked on Verify code, you will see a notification, that your email ID is verified.
7. After Email ID is verified, kindly fill in the necessary details and click "**Create**", your iX Hello account will be created and click continue.

<figure><img src="/files/hRoVWoG5YoV1xuArH6Ln" alt="" width="296"><figcaption></figcaption></figure>

<figure><img src="/files/nDSzWYM91oAbtm9I1RSy" alt="" width="296"><figcaption><p>, </p></figcaption></figure>

8. Once you click continue you will be redirected to the login page, kindly use your email id and newly created password to login.

<figure><img src="/files/I59ewciIwV8NNYZ6g9Qx" alt="" width="563"><figcaption></figcaption></figure>

9. After creating the account and logging in, you will land on the iX Hello screen as shown below:

<figure><img src="/files/R09qqH9eOlcQIRq3Dgpv" alt=""><figcaption></figcaption></figure>


# Unified iX Navigation

#### Overview

Unified iX Navigation is a cohesive and consistent navigation framework designed to work seamlessly across various applications and tools within an enterprise digital platform.

#### Objectives

* Reduce Cognitive Load: Simplify the user experience by providing a uniform navigation structure.
* Enhance Discoverability: Make it easier for users to find what they need across different modules and workspaces.
* Enable Seamless Workflow: Allow effortless transitions between tasks, regardless of the product or module being used.

#### Key Features

* Global Header/Shell: A single, universal header that remains constant across all applications.
* Consistent Navigation Patterns: Utilize the same menus, search functionalities, and favorite settings throughout.
* Unified Mental Model: Establish a common understanding for users across all experiences.

This unified approach is essential for modern experience-led platforms (iX platforms), promoting efficiency and user satisfaction.

#### Unified Product Navigation

We are introducing an improved navigation experience that allows users to seamlessly move between products within the same environment.

When logged into the staging environment of a product (e.g., Customer), you can navigate to the staging environments of other products, such as:

* Employee
* Customer V2
* Reporting
* Demo
* Knowledge

Navigation options are shown only for available products in the current environment. Unavailable products or environments will be disabled or grayed to prevent confusion.

<figure><img src="/files/d09aTCK1mUdmPAadS23n" alt=""><figcaption></figcaption></figure>

#### Simplified Navigation Experience

Currently, navigation to other products appears as multiple links in the secondary navigation, which can feel cluttered.

To improve usability:

* We are reducing multiple links into a single click per product
* Clicking a product (for example, Employee) will take you directly to:
  * The correct staging environment
  * The appropriate regional instance, based on your current region
* Admin links are being removed from this navigation to keep the experience focused and cleaner

<figure><img src="/files/YF6Q2ft5f0h778LyAEg7" alt=""><figcaption></figcaption></figure>

#### Unified Environment Consistency Across Products

This update aims to ensure a seamless transition between products without altering environments:

* If you're in Staging, any other product you select will also open in Staging.
* This consistent behavior is guaranteed across all supported regions.

#### **Login Experience: Current vs. Future**

Current Behavior:\
Switching between products often requires logging in again.

Future Improvement:\
A unified login experience will enable seamless access across all products without repeated logins.


# UI/UX Refresh (Aligned with iX Suite Design)

#### **Overview**

As part of our ongoing platform evolution, iX Hello Customer is progressively aligning its user interface and experience with the broader iX Suite design system. This ensures a more consistent, intuitive, and unified experience across all sub platforms, including:

* Employee (v1 & v2)
* Customer (v1 & v2)
* Knowledge (upcoming)
* Reporting (expected alignment)

These enhancements are being rolled out incrementally, with each release introducing targeted improvements.

**1. Unified Login Experience**

The login page has been redesigned to match the Employee and V2 platform experience, providing:

* A consistent visual layout across platforms
* Improved usability and accessibility
* A cohesive brand experience aligned with iX Suite standards

<figure><img src="/files/g58Ig7S2M5S0bEGli8H1" alt=""><figcaption><p>iX Hello Employee</p></figcaption></figure>

<figure><img src="/files/CErr6WCZwH3S6kIvJvYc" alt=""><figcaption><p>iX Hello Customer v2</p></figcaption></figure>

**2. Introduction of the “Wafer Grid” Navigation**

A new wafer grid icon has been introduced to enable cross platform navigation.

**Key capabilities:**

* Acts as a central launcher for iX Hello products
* Provides quick access to:
  * Employee
  * Customer V2
  * Knowledge
  * Reporting
  * (Coming soon) iX Demo
* Enables seamless switching between platforms without:
  * Closing the browser
  * Manually navigating to different URLs

<figure><img src="/files/mEyXzhc73IjqASQ3nAJw" alt="" width="366"><figcaption></figcaption></figure>

**3. Unified Access**

The wafer grid is part of a broader move toward unified login and navigation across the iX Suite.

* Users can easily jump between platforms from a single entry point
* Currently, users may still be prompted to reauthenticate when switching platforms
* Future updates will further streamline authentication for a truly seamless experience

More information on Unified Navigations refer to [this](/ixhc/general/unified-ix-navigation)

The iX Hello Customer UI refresh marks a significant step toward a unified iX Suite experience. By standardizing design and simplifying navigation, we are enabling users to move more efficiently across platforms while maintaining a consistent and cohesive interface.


# iX Hello: Basic Mode

Assemble and use your first digital intelligent assistant without any code

## Introduction

Let’s discuss an idea. I would love to have an AI assistant for my business, that would handle my customer requests whether through a chat, SMS, or call. I also want the assistant to be ready as quickly as possible since there is a bottleneck in my customer service and it’s impacting my business.

iX Hello ‘Create My Assistant’ could be the perfect solution to this scenario. GPT-Service Assistant 3.1 is an intelligent assistant powered by Generative AI that can be trained to support any business. Gather information on how you would want your customer representative to present your business, feed that information to iX Hello GPT Service Assistant, and voila! You have your CR ready within minutes. Now test and train the assistant for more precise and accurate information as you would like.

**What iX Hello GPT Service Assistant is?**

It is an intelligent assistant powered by Generative AI. It is a template that can be trained to behave in the way you define.

**What can I expect from it?**

It can respond to your customers via chat, SMS, or call. It also supports digital human and digital human on Kiosk mode, providing the users with a human interaction like experience rather than just chat.

**Is programming knowledge required for this?**

No, you just give information on how you want to present your representative. It feeds on your input to mimic the behavior, speaking style, and personality. It answers customer queries based on the data sources you’ve asked it to refer to.

**Will I be able to track who interacted with my assistant?**

Yes, you can get a quick summary of a number of interactions, the number of interactions today, the total follow-ups as well as the chat history. You could use these interactions to determine whether the assistant is giving desired responses. And if not, you can use these metrics to better train it.

Now, let’s look at the step-by-step process of how you can prepare a fully functional assistant within a matter of minutes. To make this process smooth, we suggest that you prepare these things beforehand.

a.     Name of the assistant (what you think would be the best identification for the bot)

b.     Your Business Details

·       Name

·       General Introduction

·       Address

·       Contact (Phone and SMS)

c.     Your Services and Products (Optional)

d.     Business Hours

e.     FAQs

f.       Your Assistant Personality (This would define the personality of the assistant. You could use examples like ‘You are very friendly to customers, you try to base your responses on facts and figures, you are not biased’ etc.) You are shaping the personality of the bot.

g.     A goodbye message

h.     Training materials: These could be websites or documents; you want the bot to use as a data pool for answering customer queries.

If you’ve gathered these, you should be able to thoroughly follow the steps below and have your first assistant ready in no time.

## Pre-requisite:

1. You need to have an iX Hello account for this. Contact our team to have your account created.
2. Login in and Click on Add a New Assistant.

<figure><img src="/files/dDZp3mIW2hTWVSSmJ91S" alt=""><figcaption></figcaption></figure>

## User Guide

1. This takes you to iX Hello Blueprint Library, which is a collection of templates that you can use to create ready-to-use AI-Apps.

<figure><img src="/files/LQ15uSOISwRnbyUML7TP" alt=""><figcaption></figcaption></figure>

2. You can explore the possibilities here. But for now, let’s focus on creating the GPT Service Assistant. On the search bar, type ‘GPT Service Assistant’.

<figure><img src="/files/pE2fiMxCf3EWOSafZbsP" alt=""><figcaption></figcaption></figure>

3. Click on Create App. Fill the details as you proceed.

<figure><img src="/files/GhsRYN7dNU9TwmiJTjNG" alt=""><figcaption></figcaption></figure>

4. Once you’ve filled in business details, click on Next to enter about your services and products.

<figure><img src="/files/H7eofDMdPspoVGQ7f86P" alt="" width="563"><figcaption></figcaption></figure>

5. Provide information on your business hours and FAQs. If you want to add more FAQs, you can click on Add New FAQ.

<figure><img src="/files/sXJ62QJjm4smmNoWZrUL" alt="" width="563"><figcaption></figcaption></figure>

<figure><img src="/files/RYymP9EriR6hO2nkDus1" alt="" width="563"><figcaption></figcaption></figure>

6. While setting the assistant details, you can describe the persona of your assistant like ‘I am friendly, I actively listen to customer queries, I provide accurate and latest information etc. This field defines the guardrails for your app. You can also configure Welcome and Goodbye messages. You can add Buttons for different services which are linked along with the welcome msg, user can simply click on the button and get the response instead of typing in the query manually. Then click Next.

<figure><img src="/files/bEabfz0T8rNIs5jOhKEY" alt="" width="563"><figcaption></figcaption></figure>

7. After these setups, another important step is training. You are providing a knowledge pool on which the assistant will base its responses. You could provide web URLs or just plain documentation.

<figure><img src="/files/9zRNe6ratfWJJH3YnnyJ" alt="" width="563"><figcaption></figcaption></figure>

8. Now click on Save, and just like that you created a working digital intelligent assistant.&#x20;

<figure><img src="/files/HR7Ga11AFq5Hu63XOCnK" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/Fdaup7vegIbmS62NLV8W" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/doopCh8VLVIDG0ZDDRMF" alt=""><figcaption></figcaption></figure>

9. You will also receive an email notification when the assistant is ready.

Now that the assistant is ready, you can use it over:

* Webchat
* Digital Avatar
* Call/ SMS

The following chapters discuss in detail how to set up the assistant for these different modes and what you can expect on these setups.&#x20;


# Webchat Assistant

#### Test Webchat

Activate the chat feature. You’ll have now option to ‘Chat Now’ and ‘Manage’. On clicking ‘Chat Now’, a pop-up window will occur that allows you to experience how the assistant interacts with your customers.

<figure><img src="/files/uxKOUuo2Iu7IDHoqWBrW" alt=""><figcaption></figcaption></figure>

No deployment required; no additional setup required; you can instantly test your assistant within iX Hello and determine if it is ready to be launched for official use.

You can obtain a direct chat URL for this assistant. Or you can find a JavaScript code in the settings which can be embedded to any of your websites to provide your personalized chat assistant to your website visitors.

**Domain Restrictions**

You can also define domain restriction rules like ‘Show widget only on specific domain and/or URLs. This facilitates security and ensures that the widget is being used on authorized sites only.&#x20;

<figure><img src="/files/kucVCsahPPlaxhnni7eg" alt=""><figcaption></figcaption></figure>


# Direct Chat Link

Now the digital assistant is ready with web chat options. You can obtain and provide a ‘Direct Chat Link’ to this assistant for anyone.

Click on ‘Manage’ on the Web chat option, you’ll get settings for Web Chat.

<figure><img src="/files/ZniTk7WA2fgx1desF0iE" alt="" width="563"><figcaption></figcaption></figure>

Click on ‘Open’ and this will open the chat assistant in the new browser window.

<figure><img src="/files/LrwCPBElbLV122BHa002" alt=""><figcaption></figcaption></figure>

Imagine all the possibilities, you could attach this direct link to your emails or share it with your customers instantly, to help with their queries.


# Direct Chat Link

Now the digital assistant is ready with web chat options. You can obtain and provide a ‘Direct Chat Link’ to this assistant for anyone.

**Note: Authentication is an optional feature which is use case based/determined.**&#x20;

### **Chat Link Authentication**

To successfully initiate a Direct Chat Link, it's imperative to secure the proper Authentication procedures. This process is vital to ensure that only authorized users can access the chat, maintaining the integrity and safety of all communications. You can easily implement this by activating the "Require Authentication" toggle button within your security settings. This step not only bolsters your overall security posture but also provides peace of mind knowing that each session is protected and verified. Proper authentication safeguards personal information, mitigates unauthorized access, and ensures compliance with privacy regulations.

When the toggle button is enabled, it ensures authorized access, complies with IT security requirements, seamlessly integrates with enterprise systems, and secures data.

### Authentication Process for Webchat

When the "Require Authentication" toggle is enabled, the following steps occur:

1. The 'Require Authentication' toggle is added to the webchat settings.
2. Enabling this toggle initiates user authentication upon accessing the bot.
3. If authentication is successful, the user can proceed to use the bot for its intended purpose.
4. If authentication fails, the user receives an 'Invalid Access' message and is denied access.
5. The authentication process is triggered whenever the bot is accessed via URL or embedded webchat.

Click on ‘Manage’ on the Web chat option, you’ll get settings for Web Chat.

<figure><img src="/files/VUHRcRL4KGtVZ1RA3Fk6" alt=""><figcaption></figcaption></figure>

Click on ‘Open’ and this will open the chat assistant in the new browser window.

Note: If you're already logged into the iX Hello app and attempt to use a direct chat link, you'll need to log out of the app first to authenticate successfully.

If you share the direct chat link with outsiders / clients, by enabling both the toggle buttons - Public access and Require authentication. Then below are the steps to follow:

Ensure both "Public access" and "Require authentication" are enabled when sharing the direct chat link with outsiders or clients. Follow these steps to proceed:

1. Enable the "Public access" toggle to allow external access.
2. Enable the "Require authentication" toggle to ensure users authenticate before accessing the chat.
3. Share the direct chat link with the intended users.
4. Instruct users to authenticate using their credentials when accessing the link.
5. Monitor access logs to track usage and access attempts.

<figure><img src="/files/zp8QTG9YNOYfcKkd5JLO" alt="" width="302"><figcaption></figcaption></figure>

Once you click on, Please Login, you will be navigated to the below screen

Enter your Email ID and click Next

<figure><img src="/files/zkGIMCIXyMdXUI0tfpEh" alt=""><figcaption></figcaption></figure>

Once clicked Next, you will be successfully launching the bot though direct chat link.

<figure><img src="/files/cASluujJN0j7tpEgDzBh" alt="" width="375"><figcaption></figcaption></figure>

Imagine all the possibilities, you could attach this direct link to your emails or share it with your customers instantly, to help with their queries.


# Embedded Chat Assistant

On the previous settings page, you can also find ‘Chat Widget Code’. You can paste this code into your website, and you’ll have already functional chat service available on your website just like that.

<figure><img src="/files/4erK5XWNOlo0Vj6AiLM4" alt="" width="563"><figcaption></figcaption></figure>

A sample representation is shown in the images below (see bottom right corner):

<figure><img src="/files/Oi47geaHkyTssmu3X9HF" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/Al6u63l89eYbKLOZLj5p" alt="" width="327"><figcaption></figcaption></figure>


# Digital Assistant

You can use Digital Avatar to make the app interaction more interesting and engaging. You can either use the combination of Digital Avatar and Chat (Digital Avatar mode) or use Kiosk Mode.


# Digital Avatar

To break the monotonous chat interactions and provide some sense of human-like interactions, iX Hello also supports Digital Human mode. This feature is also suitable for those who prefer hands-free voice interactions; eliminating the need to type and read everything.

To use this, click on Activate for ‘Digital Human’.&#x20;

<figure><img src="/files/B2EhaEDH6xAQQxRJiCn5" alt=""><figcaption></figcaption></figure>

Then Enable Digital Human Feature.&#x20;

<figure><img src="/files/PA486ADf8mP25trPdPY7" alt=""><figcaption></figcaption></figure>

iX Hello supports a bunch of Digital Human personas, you can select what fits the best for you. Determine the voice and language.

You can even select the working voice and speaking style. Provide a sample text and test how the digital human sounds.

<figure><img src="/files/y036ewxvAiR3ItHCvtoV" alt=""><figcaption></figcaption></figure>

You also have options to control whether you want to show this digital human on the chat screen.

Audio/ Mic controls can also be controlled on the chat screen.

Once you’ve set this up, you can click on Save and test the Digital Human in chat.

<figure><img src="/files/68DfGZPqwi3rMa1WJdLX" alt=""><figcaption></figcaption></figure>

Click on Chat Now. Your previous chat screen will now look something like this.

<figure><img src="/files/OSzt6b8Q1DOuUrf0O2Az" alt=""><figcaption></figcaption></figure>


# Kiosk Mode

Another interesting feature you can use is the use of Kiosk controls. Click on Enable Kiosk controls to activate this. Now save and Chat, your chat screen now appears like this.

<figure><img src="/files/MFcbFANm3mm2KWym77xt" alt="" width="563"><figcaption></figcaption></figure>

You can Tap to talk and interact with the assistant.

In these few simple steps, you can have a personalized, customized assistant in no time.

For customers that are available only via call/ SMS, you can opt for Call/Text feature.


# Call/SMS Assistant

### Call / Text

iX Hello Gen AI assistant can be set up to respond to phone calls/ SMS. It supports call/SMS feature based on Twilio.&#x20;

To set up the assistant for Call/ Text settings, click on Activate.

<figure><img src="/files/fJpfVaqEQgY6KRdMFo3x" alt=""><figcaption></figcaption></figure>

Enable Twilio bot use for the App.

<figure><img src="/files/5IeVSXh9SSafy2eNLQrG" alt=""><figcaption></figcaption></figure>

The next step is to assign Twilio phone numbers to the App. \
a. Select Country (available options US / Canada)

b. Select Scheme (Local or Toll-Free). For the Toll-Free number, you've to make additional setups in the Twilio account.

c. Twilio Phone Number: The list provides available Twilio Phone numbers from which you can choose any that you prefer. Once you select and save, the particular number will be occupied by this assistant and will only be freed again once you disable Twilio Bot for this app.

d. Voice Gender: You can also configure the voice gender that will respond to calls on this number. For example, in areas where a female representative is conventionally expected, you can set the voice Gender to female and vice versa.

e. Voice Profile: You can also select a voice profile that you feel like matches the personality of the assistant. **Enhancements are done using Real Time Voice from Eleven Labs Voices**.

Benefits: Availability of more natural sounding voice libraries

• Reduced response processing

• Improved latency in voice interactions

• Better natural interaction experience

f. To check if you like the Voice of the assistant, you can click on the Play button that will play the sample text in the Voice Gender and Voice Profile you just selected.

<figure><img src="/files/s57VaYEZUc2iemd7IkLL" alt="" width="563"><figcaption></figcaption></figure>

Use the "Select Language" drop-down to choose different languages,

<figure><img src="/files/I5fp6sPcwcJjrCFeXSK1" alt=""><figcaption></figcaption></figure>

Now, click on Save and your assistant will now readily respond to calls and SMS from your customers.

You can manage the Twilio settings by clicking on the Manage button.

<figure><img src="/files/HlIJgDvD8sBJprJLTnSn" alt=""><figcaption></figcaption></figure>

## Deactivate Call/SMS Assistant

To deactivate the Call/SMS feature for the assistant, click on the Manage button. Then, turn off 'Enable Bot to use this App publishing' and click on Save.

<figure><img src="/files/ITtgMdO0ovQS9gMwCoh1" alt=""><figcaption></figcaption></figure>

The assistant now no longer be available for call/ SMS queries and the Twilio number assigned to this assistant will be added to the list of available phone numbers in Twilio as well.


# Administrative Purpose

iX Hello enables you to view chat history and logs of interaction with the assistant. You can view a summary of the total number of interactions, the interactions today, and total follow-ups.

<figure><img src="/files/aWDPJXDiKvlaJswH6XY3" alt=""><figcaption></figcaption></figure>

Chat history can be accessed from the top right menu.

<figure><img src="/files/qDyKOsAxdCemRFF1VPsw" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/mpT7TkWZsRyXp2DjdasK" alt=""><figcaption></figcaption></figure>

You can have multiple assistants deployed. Hence, you can filter chat history by your assistant and by date range.

By now, we hope you’ve successfully created and tested your digital assistant. Feel free to reach out to iX Hello at <https://www.ixhello.com/> to discuss more possibilities.


# WebCrawl

### Overview:

A web crawler, also known as a "spider" or "bot," is an automated program that systematically browses the internet to discover, index, and store information from webpages. Used mainly by search engines like Google, it builds a massive, searchable index of the web, helping users find relevant content quickly.

### Prerequisites:

* **An iX Hello Account** For steps to [create an iX Hello account,](https://app.gitbook.com/o/-M8Qw0HjmL3rDRYUXBX0/s/-M8XHvUsfyTUFLvToHqD/general/ix-hello-create-account) follow the steps at
* Use this link to [Sign up for iX Hello platform](https://bots.ixhello.com/Account/Login?ReturnUrl=%2F).
* To create an APP using Quick Start App refer to [this](https://app.gitbook.com/o/-M8Qw0HjmL3rDRYUXBX0/s/-M8XHvUsfyTUFLvToHqD/general/rag-file-upload/file-upload-in-studio-mode-1/quick-start-apps)

### **How to create an Assistant using WebCrawl in iX Hello:**

1. On the Home Screen, click on the Quick Start App Button.

<figure><img src="/files/2b154KSwmLdaKE86DMxt" alt="" width="563"><figcaption></figcaption></figure>

2. You will be navigated to Quick Start Apps screen and then click on "Add a New Assistant" button.

<figure><img src="/files/ZHJpM5dUJMuGPrUnfi8f" alt="" width="563"><figcaption></figcaption></figure>

3. You will see the Quick Start Apps Library, which helps you to create an app using ready to use AI apps.
4. To create a new assistant using the Add a New Assistant button and choose Quick RAG-App.

<figure><img src="/files/cUUGwq6FfiHTxMSplPCx" alt="" width="563"><figcaption></figcaption></figure>

5. Fill out the Configuration screen of General Information Tab.
6. The table below explains each of the inputs shown on this screen.

| Input                         | Information                                                                                                                                                                                                                                                                                                    |
| ----------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Name                          | <p>This is the display name of the assistant.</p><p> For Ex: WebCrawl Test                  </p>                                                                                                                                                                                                               |
| Description                   | <p>Explains what the assistant is designed to do.</p><p>For Ex: WebCrawl Test</p>                                                                                                                                                                                                                              |
| Welcome Message               | This is the first message the assistant sends when a conversation starts.                                                                                                                                                                                                                                      |
| Goodbye Message               | This is the closing message when the conversation ends.                                                                                                                                                                                                                                                        |
| Persona and Behavior (Sample) | <p>Defines the assistant’s tone, personality, and response style.</p><p>For Ex: I am a concise and focused assistant, providing accurate information based solely on the provided Content files from URL. I aim to deliver clear and relevant answers to your queries from Content files from URL provided</p> |

<figure><img src="/files/k5ukAYvcfnVeSNGDtuK1" alt="" width="563"><figcaption></figcaption></figure>

7. Fill out the Configuration screen of Knowledge Base Data Tab,

| Input                     | Information                                                                                                                                                                                                                                                                                                                                                                                               |
| ------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| SharePoint Upload Section | Displays files and folders from the selected SharePoint site.                                                                                                                                                                                                                                                                                                                                             |
| Instructional Text        | <p>Connect to SharePoint, browse folders/files, and select content for <strong>AI training</strong>.</p><ul><li><strong>Supported File Types</strong>: <code>.pdf, .docx, .csv, .xls, .xlsx, .txt, .ppt, .pptx, .gif, .jpg, .jpeg, .png, .webp</code>.</li><li><p><strong>Upload Limits</strong>:</p><ul><li>Max per file: <strong>200MB</strong></li><li>Total: <strong>1GB</strong></li></ul></li></ul> |
| Action Buttons            | Click Save Button to save all the information updated.                                                                                                                                                                                                                                                                                                                                                    |

<figure><img src="/files/RyVKSoFCrhtOo2s9baHp" alt="" width="563"><figcaption></figcaption></figure>

8. The bot created will appear under My Assistant and enable the PII Redaction by clicking on the warning "Click Here" and hit the Save button on the pop-up.
9. To know more on PII Redaction refer to [PII-Management](/ixhc/general/pii-management)

<figure><img src="/files/Up12GnP3bSKjJoe7OLaT" alt="" width="563"><figcaption></figcaption></figure>

<figure><img src="/files/3FLNwWBWTpQHuwzlPFRW" alt="" width="563"><figcaption></figcaption></figure>

10. Click on "Activate" button to activate the bot. You can click on "Chat Now" and start the conversation with the bot created.
11. Based on the content in the file, which is uploaded, you can query the bot accordingly:

<figure><img src="/files/73MwnF9VnHK8Vs3Jc8Rp" alt="" width="563"><figcaption></figcaption></figure>

12. You can access the same bot through Custom Apps.

{% hint style="info" %}
Using the Quick Start App to create this assistant ensures that all the necessary tabs in custom apps are automatically generated or updated. This includes the AI-Content tab, Methods Tab, Intents, Utterance, Slot, and Input.
{% endhint %}

<figure><img src="/files/NTWWI34h8NghREJyNeig" alt=""><figcaption></figcaption></figure>

13. The list of contents provided while creating the assistant are as below

<figure><img src="/files/H9mnx1NYmy9OHCbzAZJn" alt=""><figcaption></figcaption></figure>

**a. Preview Content**

* Let's you view the data that has been extracted or crawled from the specified URL.
* Helps verify that the correct content was fetched.
* Ensures the data is relevant and clean before using it for AI training.

**b. Refresh Content**

* Updates the stored content by re-fetching data from the original URL.
* Keeps training data current when the source website changes.
* Ensures the AI model uses the latest information.

**c. Delete**

* Removes the selected training URL and its associated content from the system.
* Cleans up outdated or irrelevant sources.
* Prevents unnecessary data from influencing the AI model.

### Let's test the assistant

<figure><img src="/files/JJXXCIJxgAJ7tDK7G8Zq" alt="" width="563"><figcaption></figcaption></figure>

When the user clicks on the provided links, the user will be able to view the content screen associated to the link provided with in the Chat Widget.

<figure><img src="/files/R7yGC3tEwXjh6LnVQLK2" alt="" width="563"><figcaption></figcaption></figure>


# iX Hello: Studio Mode

## Objective

The purpose of this document is to describe how to create a generative AI app in the studio mode in iX Hello. After completing the following series of documents, you should be able to create a generative AI app for web chat and phone/text.

a. [Prerequisites](/ixhc/general/ix-hello-studio-mode/prerequisites)

b. [Steps to Build Custom App](/ixhc/general/ix-hello-studio-mode/steps)


# Prerequisites

Pre-requisite to complete before you start custom app creation using Studio Mode

## Integration with Gen AI

In order to create the data source for the app, you need to create an integration with Gen AI. Follow [this ](/ixhc/general/enterprise-connections/generative-ai)document to create an integration with Generative AI.

## Integration with Internal Content

You can manage your content like text, audio, and other media files within iX Hello. They can later be used in custom apps. Following through [this ](/ixhc/general/enterprise-connections/internal-content)short document, you should be able to readily connect an integration with iX Hello Internal Content.

## Create content&#x20;

You can prepare any text, media, html files to use as content for the app.&#x20;

Or you can use this sample file to proceed with the process. Download this 'demoContent.json' file here.

{% file src="/files/HN3CXv7MCZBYNDcVDDba" %}

It is a JSON file that contains the basic information of a business-like description, opening hours, contacts, etc. &#x20;

Now, on the iX Hello portal, on the left navigation, go to Content -> Content and click on Add New Content.

<figure><img src="/files/xQseZrDZbKEsLI8SXmbP" alt=""><figcaption></figcaption></figure>

Provide the following details:

Content Id: A unique name for your content e.g. DemoContent.

<mark style="color:red;">Note</mark>: Save Content Id for use further in the process of creating a custom app.

Content Name: A unique name for your content. e.g. DemoContent

Category: Default

Text: Sample content information

SSML: False

HTML: False

Media or Data: True

Media Type: JSON

Media Source: Upload File

Media File: Select the downloaded file 'demoContent.json'.

<figure><img src="/files/f7Ex9wfaaYRK1vWlu6z1" alt=""><figcaption></figcaption></figure>

After filling in the above details, click on Add, and the content should be ready to use.

<mark style="color:red;">Note</mark>: Make sure that you've selected 'Default' as a category. Contents in iX Hello are categorized by their category and only those apps that have the same Category as the content can access these contents.


# Steps

Creating custom apps requires you to follow few sequential steps.&#x20;

a. [App Creation](/ixhc/general/ix-hello-studio-mode/steps/app-creation)

b.[ Data Source Configuration](/ixhc/general/ix-hello-studio-mode/steps/data-source-configuration)

c. [AI Content Setup](/ixhc/general/ix-hello-studio-mode/steps/ai-content-setup)

d. [Method Setup](/ixhc/general/ix-hello-studio-mode/steps/method-setup)

e. [Intent Setup](/ixhc/general/ix-hello-studio-mode/steps/intent-setup)

f. [Slot Setup](/ixhc/general/ix-hello-studio-mode/steps/slot-setup)

g. [Utterance Setup](/ixhc/general/ix-hello-studio-mode/steps/utterance-setup)

h. [Input Setup](/ixhc/general/ix-hello-studio-mode/steps/input-setup)

i. [Input Parameter Setup](/ixhc/general/ix-hello-studio-mode/steps/input-parameter-setup)

j. [Publish Setup](/ixhc/general/ix-hello-studio-mode/steps/publish-setup)

k. [Test](/ixhc/general/ix-hello-studio-mode/steps/test)


# App Creation

1. Login to iX Hello and navigate to **Apps** and **Custom App**. Click on **New Custom App**.

   <figure><img src="/files/37BBf4Zn7VRamryA9MQ0" alt=""><figcaption></figcaption></figure>
2. The Add Custom App window opens. Enter the App details as shown below. Make sure the **This is a Generative AI App** checkbox is ticked.

   <div align="left"><figure><img src="/files/9fCcsfhRtiFx8NVz0tcp" alt="" width="375"><figcaption></figcaption></figure></div>
3. Click **Save** and the app will appear on the Custom App list. Click on the app name or the ellipses to open the details.

   <figure><img src="/files/kbj2PXpJnoS1qO5WfaDl" alt=""><figcaption></figcaption></figure>


# Data Source Configuration

1. Once you click on the App name, it will take you to the Data Source tab. <br>

   <mark style="color:red;">Note:</mark> Before proceeding, make sure that you have all prerequisites to follow through the next steps smoothly. Kindly refer to [this link](/ixhc/general/ix-hello-studio-mode/prerequisites) to set all your prerequisites.<br>
2. Open the **Data Sources** tab and click **Create New Data Source**.

   <figure><img src="/files/3x0FbdmDw1sAXpmQFHtF" alt=""><figcaption></figcaption></figure>
3. Give the DataSource a Name. From the dropdown list, select the integration to Generative AI created in the Prerequisites. Click **Save**.

   <div align="left"><figure><img src="/files/Otyr36yvpJJZalu00mYd" alt="" width="333"><figcaption></figcaption></figure></div>
4. Next, repeat steps 2 and 3 to create another DataSource similar to the Generative AI data source, except select the integration to Internal Content instance that you created in the Prerequisite section.

   <div align="left"><figure><img src="/files/5mVc7bKDWh9OOx4KQanR" alt="" width="375"><figcaption></figcaption></figure></div>
5. Now all the Data Sources are ready and available in the list as shown below.

   <figure><img src="/files/JC6tIMDtJ90kJ2xRcb6b" alt=""><figcaption></figcaption></figure>

Note: You can find other Data Sources&#x20;


# AI Content Setup

1. &#x20;The AI Content tab manages AI contents for the app. For user convenience, as soon as any 'Generative AI' app is created, sample AI Contents are created for the app as follows.&#x20;

   In order to create new AI Content, click on “Add New AI Content”.

<figure><img src="/files/A4C6hS7Izz5OyQao6EEN" alt=""><figcaption></figcaption></figure>

2. Follow the steps [here](/ixhc/general/ai-contents/add-ai-content) to add new AI-Content.&#x20;
3. Once the AI-Content is created, you can Preview, Refresh, or Delete the content from this interface.


# Method Setup

Our next step would be to set up methods. Methods detail how the app is going to handle user requests.&#x20;

Depending upon the type of data source, e.g. Gen AI, Twilio, internal content, etc., different operations can be carried out.&#x20;

For instance, with a Twilio data source, the user might be able to make calls, send SMS, etc., but with a Gen AI data source, the user might be able to get responses to queries in chat.&#x20;

Such different operations are termed 'Request Types'.&#x20;

For now, let's construct methods for our 3 different data sources from the previous step.

#### Method: Generative AI

1. Go to “Methods” tab, select “Generative AI” Data Source and click “Add New Method” as shown below.

<figure><img src="/files/pjp2e89LLDfLUkCVRFwj" alt=""><figcaption></figcaption></figure>

2. For “Methods Input” tab, select the Request type and subsequent details as shown below.

<figure><img src="/files/ABpLcaaANK8YOlyvKIMH" alt=""><figcaption></figcaption></figure>

3. For “Method Output” tab, use the details as shown below

<figure><img src="/files/fGsbYIDndL0yGVbc95a7" alt=""><figcaption></figcaption></figure>

4. For “Method Result” tab, enter the details as shown below

<figure><img src="/files/VnkVj8xsZbkOdSKolifg" alt=""><figcaption></figcaption></figure>

With this, you've now set up for the Gen AI chat responses.&#x20;

You might also want to provide more specific information about your business. So, for that purpose let's use the content that you created in the prerequisite session and create a method that reads from that content.

#### GetBusinessInfo Method

Change the data source to Internal Content in the dropdown and click on Add new Method.

<figure><img src="/files/vtmX2FdvP60Mmak8HeWI" alt=""><figcaption></figcaption></figure>

Add following details to create a method.

| Tab                | Parameter             | Value                                                        |
| ------------------ | --------------------- | ------------------------------------------------------------ |
| Method Inputs      | Method Name           | GetBusinessInfo                                              |
|                    | Request Type          | GetJSONContents                                              |
|                    | Parameters: ContentID | The name of the content that contains your business name     |
|                    |                       | e.g. DemoContent                                             |
| Method Output      | Result Type           | ListOfRecords                                                |
|                    | Record Limit          | 10                                                           |
| Method Result      | Record JsonPath       | $.\[\*]                                                      |
|                    | Valid Result Template | {{Records}}                                                  |
|                    | Record Template       | {{Session.StoreDescription}}                                 |
| Session Attributes | SessionAttributes     | Name : StoreDescription, Value: {{Results.StoreDescription}} |
|                    | SessionAttributes     | Name : PhoneNumber, Value: {{Results.PhoneNumber}}           |
|                    | SessionAttributes     | PName : Email, Value: {{Results.Email}}                      |

<figure><img src="/files/VBytZWy8xHFjTDN3sYey" alt=""><figcaption></figcaption></figure>

In order to get the value for {{Results.StoreDescription}}, {{Results.PhoneNumber}}, and {{Results.Email}}, you need to define them as Managed Results. You can find the StoreDescription, PhoneNumber, Email etc on the JSON Content file you uploaded in the prerequisite section.

To add Managed Results, once the method is created, click on the Action button and Select 'Manage Results'

<figure><img src="/files/gichlWiU4ZE95NC4vTUk" alt=""><figcaption></figcaption></figure>

Enter the following details. By doing this, you are reading the info from the content file into these variables for use in the method.

<figure><img src="/files/kb7zLxlk6dDQlwlxKNQ5" alt=""><figcaption></figcaption></figure>


# Intent Setup

1. Go to the “Intents” tab and click “Create New Intent”.  Intents are the user's reasons for interacting with the app. Hence, let us create intents for the custom app.

#### &#x20;Chat Intent

<figure><img src="/files/HRlhtxyYXmwcUhlxbVaj" alt="" width="386"><figcaption></figcaption></figure>

For more information related to the BeforeAfterMethod feature see [here](/ixhc/apps/intentfeature-beforeaftermethod).

At the end of this step, your intent list should look something like this.&#x20;

<figure><img src="/files/yBNrS42kQ8PTwMcA6xnK" alt=""><figcaption></figcaption></figure>


# Slot Setup

When interacting with intents, you expect inputs from the users. Those inputs are mapped to slots. You need to capture those inputs for processing the intent.

Hence, slots are created for each intent that expects a user input.&#x20;

On the Slots tab, select the Intent you are creating the slot for in the Select Intent drop down.

Then click on Create New Slot.

a. Chat Intent

<figure><img src="/files/2fwMo1Oe0cvPw3Kuezga" alt="" width="374"><figcaption></figcaption></figure>

<figure><img src="/files/xKOsd8kPdtMPAR3sSGFA" alt=""><figcaption></figcaption></figure>

Enter the details and click Save. Your slot should have been created from ChatIntent.


# Utterance Setup

Utterances are the keywords that trigger intents. i.e. when the users say these words, the app knows that the user is looking to initiate a particular purpose/conversation with the app.&#x20;

You can define single or multiple utterances for an intent.&#x20;

Similar to slots, an utterance also corresponds to a single intent.&#x20;

An intent might have multiple utterances, but each utterance is mapped to only a single intent.

1. Go to “Utterances” tab select “ChatIntent” as click on “Create New Utterance” as shown below

<figure><img src="/files/nPuz9ikafvbmSCVKmiEH" alt=""><figcaption></figcaption></figure>

1. Enter the Utterance details and click “Save” as shown below

<figure><img src="/files/No3Awi9RljlyQiKCmopZ" alt="" width="409"><figcaption></figcaption></figure>


# Input Setup

1. Navigate to the “Input” tab and click “Create New Input” as shown below

<figure><img src="/files/m83Pe7YLlCI6XC6sXy3D" alt=""><figcaption></figcaption></figure>

2. Enter the details as shown below and click “Save”

<figure><img src="/files/nUcPdRjF5joNtWIKen8E" alt="" width="424"><figcaption></figcaption></figure>

<figure><img src="/files/6tlK3ylV1XGHKtpJyP0D" alt=""><figcaption></figcaption></figure>


# Input Parameter Setup

1. To add input parameters, you will find a "Manage Parameters" button under Input. Click "Manage Parameters."

<figure><img src="/files/ISzEhfWq0Mkxti7KdwDK" alt=""><figcaption></figcaption></figure>

2. This will navigate to a different page where you can add input parameters. Click "Input Parameter" button show below.

<figure><img src="/files/PEpb0DzrmzzwBIdvOYrb" alt=""><figcaption></figcaption></figure>

2. Fill the add input parameter form as show below. Click "Save".

<figure><img src="/files/7eXgUwaAHCUpoOxTUWT7" alt="" width="403"><figcaption></figcaption></figure>


# Publish Setup

1. Finally, navigate to the "Publish" tab and click the "Add App Publishing" button.

<figure><img src="/files/Xq1d9JJRUExkDZkj08fk" alt=""><figcaption></figcaption></figure>

1. A modal with a form will pop up as show below. Fill up all the required values in the form. Importantly, select channel as "Generative AI". Click "Save."

<figure><img src="/files/sHs8bNfKBbP2oHGmuLSb" alt=""><figcaption></figcaption></figure>

This will create a new publishing profile with the provided information for an app.


# Test

iX Hello allows you to test your app without deployment hassles. Once you have created the Publish Profile, click on the Action menu on the far right. Select "Test".

<figure><img src="/files/mCQgqjouuKUGYxsbqiHj" alt=""><figcaption></figcaption></figure>

A simulated chat bot allows you to test your app.

<figure><img src="/files/o2v776ZALw6GJl96QeC0" alt="" width="327"><figcaption></figcaption></figure>

The app derives from the training data you provided so you can test and update your training contents to make your app best represent you.


# Enable Call /SMS

You can make this assistant to respond over phone call/SMS with a small change in the settings. Similar to the Basic Mode, Gen AI assistant provides call/SMS support over Twilio. On the App Publishing, you can see a flag for 'Enable Twilio Bot to use this App Publishing'.&#x20;

<figure><img src="/files/BMtjpxkcmDViyvp4GvSc" alt=""><figcaption></figcaption></figure>

Turn on the 'Enable Bot to use this App Publishing' toggle button.&#x20;

<figure><img src="/files/o0nIal0Z2FShavyqOhJe" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/U1PsDPfEFsq9EQVovOpZ" alt=""><figcaption></figcaption></figure>

You'll be prompted to setup basic Twilio setups.&#x20;

a. Select Country ( available options US / Canada)

b. Select Scheme ( Local or Toll Free) . For Toll Free number, you've to make additional setups in Twilio account.

c. Twilio Phone Number : The list provides available Twilio Phone numbers from which you can choose any that you prefer. Once you select and save, the particular number will be occupied by this assistant and will only be freed again once you disable Twilio Bot for this app.

d. Voice Gender: You can also configure the voice gender that will respond to calls on this number. For example, areas where a female representative is conventionally expected, you can set the voice Gender to female and vice versa.

e. Voice Profile: You can also select a voice profile that you feel like matches the personality of the assistant.

f. To check if you like the Voice of the assistant, you can click on the Play button that will play the sample text in the Voice Gender and Voice Profile you just selected.

Now, click on Save and your assistant is now readily responding to calls and SMS from your customers.


# Enterprise Connections

Enterprise connectivity streamlines business processes, reducing latency and improving the speed of data transfer across the network. This leads to enhanced productivity and efficiency, as employees and customers can access and share information quickly and collaborate in real time, regardless of their location.


# Generative AI

If you want to get hands on creating a Gen AI app without having to handle the hassle of setting / maintaining third party connection to providers like Open AI/ Anthropic / Azure Open AI etc., then you can use Generative AI integration provided by iX Hello.

On the left navigation, go to Integration -> Connect System.

<figure><img src="/files/UcAo35axlZve9bUjGK8y" alt=""><figcaption></figcaption></figure>

The resulting page lists all the third party and internal connections your account has.&#x20;

Click on Add New Connection.

On the resulting pop up, under 'Choose Connection' Integration Category select 'Generative AI'. Click on Next.

<figure><img src="/files/seEObgoh2oOZlM8ZAQml" alt=""><figcaption></figcaption></figure>

Provide an instance name and click on Save Settings.&#x20;

<figure><img src="/files/D2VDdW48ThvFj22F7FQX" alt=""><figcaption></figcaption></figure>

An instance of connection to Gen AI is now available and can be used as data source for your apps.

<figure><img src="/files/JQhyYygKC4Z6ngTIJtr4" alt=""><figcaption></figcaption></figure>


# Internal Content

iX Hello allows you to save your contents like text, image, audio/ video, HTML, SSML, JSON files etc. within the platform as well. They are referred to as Internal Contents.&#x20;

To access these contents from your app/ bot, you need to create an instance of connection to 'Internal Content'.

From the left navigation, go to Integrations -> Connect System and then Add New Connection.

<figure><img src="/files/FEjgPBgrcPYSTL3QVUvf" alt=""><figcaption></figcaption></figure>

Click on Add New Connection.

On the resulting pop up, under 'Choose Connection' Integration Category select 'Internal Content'. Click on Next.

<figure><img src="/files/ivhumW7QLOHKpIrUDMpq" alt=""><figcaption></figcaption></figure>

Provide an instance name and click on Save Settings/Next.

<figure><img src="/files/PFcZL2pJEHvdFLkDo2pm" alt=""><figcaption></figcaption></figure>

Upon successful connection, you can find your integration instance to Internal Content, and this now can be used as a data source for your app/ bot.

<figure><img src="/files/exLut7BwAB4RnBrkvGS7" alt=""><figcaption></figcaption></figure>


# Microsoft Office 365

Follow the steps below to setup the iX Hello integration to your Microsoft Office365 instance.

### Step 1: Obtain Microsoft Azure subscription.

To use Microsoft Office365 integration with iX Hello, it is required to have Microsoft Azure subscription. If you do not have this, get it with help of your Microsoft Office 365 administrator.

![](/files/q7lPMAtNaUoevJqHhS3o)

**Step 2: Obtain your Microsoft Office 365 user login details**

User login to Microsoft Office 365 / Azure requires username and password. iX Hello will require the same credentials to connect your voice platforms to Microsoft Office 365 in **Step 6**. Get your Microsoft Office 365 username and password from the administrator of your account of Microsoft Office 365.

**Step 3: Configure Microsoft Azure Portal by creating new App Registration for** iX Hello&#x20;

1. Launch Microsoft Azure portal by navigating to <https://portal.azure.com/#home>. If you are not logged in, you will be prompted to login by providing Azure / Office 365 credentials.

![](/files/DB7CZ9oD0VxvSW1edV5y)

1. Once you are logged in, find Azure services and click on App registrations section as shown below.

![](/files/ikPIVFfcGhWIX6YkAzTT)

1. Once you are on App Registrations screen, click on “New registration” as shown below

<figure><img src="/files/AoigD8G3pd5QAl00y868" alt=""><figcaption></figcaption></figure>

2. Fill the details as shown in screenshot.

<figure><img src="/files/5syw8OKx8b0L9IOkP4qM" alt=""><figcaption></figcaption></figure>

Redirect URI is shown as optional in the screen, however, to make this work with iX Hello , we need to provide correct URI.

Enter [https://bots.ixhello.com/](https://bots.ixhello.com/Account/Login?ReturnUrl=%2F) as the redirect URI. Make sure web is selected in the dropdown.

<mark style="color:red;">Note</mark>: The redirect URI is case sensitive.

Click on Register button.

## Application (Client) ID, Tenant ID, Client Secret, Permissions & Authentication

1. To be able to connect to Office365 from the iX Hello platform, one application registration is required, and user should have permissions to view and execute Office365 functions. Once the application registration is successful, you can copy the information as described in the steps below:
2. User will need to visit the newly created application and note the following information as highlighted with red marker. The user can find Application ID and Tenant ID market inside red contour.
3. At this stage, Azure Application is created, and you will be redirected to the Azure Application Overview page.

Copy Application (client) ID - we will need this later in iX Hello configuration

<figure><img src="/files/wPwIrD8SADT7yr5SHd4Q" alt=""><figcaption></figcaption></figure>

3. Now from left menu, click on Certificates & secrets, it will open Certificates & secrets page, on this page, click on + “New client secret” button.

<figure><img src="/files/u5QJs1iD7i9yf2FOHPk8" alt=""><figcaption></figcaption></figure>

4. You will get a small pop-up to enter description, provide description of your choice. Make sure to choose “Never” and then click on “Add” button

<figure><img src="/files/36aVmAJn3PtWEaKiQb3A" alt=""><figcaption></figcaption></figure>

&#x20; App expiration can be customized as shown below:

<figure><img src="/files/C2AXr4h0nfYHO0CHLEq6" alt="" width="375"><figcaption></figcaption></figure>

1. After user clicks “Add” button, the application secret will appear as shown below. Copy the value and store it in a safe location. This information can only be seen once, and it will be necessary to create a new secret if information is lost.

<figure><img src="/files/n78TiX2eNConZNYIlT60" alt="" width="563"><figcaption></figcaption></figure>

If you miss it, you will have to create a new secret.

1. User should now have Client ID, Tenant ID and client secret.
2. User will now need to create app permission. On left navigation menu, click “API permissions”, then click “Add a permission” button as shown below:

<figure><img src="/files/VyLEEepCps3JyKOHt4b2" alt="" width="563"><figcaption></figcaption></figure>

You can see that by default *User.Read* permission is listed. – we need to add more permissions.

1. You should see below “Request API permissions” screen, make sure “Microsoft APIs” is selected and scroll down and find “SharePoint” and click on it.

![](/files/iwu086cUCqjVXt29BBmE)

1. Now you will get permission selection screen. Select “delegated permissions” as types of the permissions and then “AllSite.Manage” permission and click on Add Permissions button.
2. Now configured permissions should look like this

   <figure><img src="/files/JJmJOrwts3DWwmqGP43c" alt=""><figcaption></figcaption></figure>

At this stage, you are done with configuration at Microsoft Azure/office 365 portal.

<figure><img src="/files/0D47Egxf9NKYuXbmZKqT" alt=""><figcaption></figcaption></figure>

### Step 4: Configure iX Hello Connection to Microsoft Office 365

1. You will need three things from Azure Portal described in previous step
   1. ClientId (explained in step 3 .e)
   2. Tenant Id (explained in step 3 .e)
   3. Client Secret (explained in step 3 .h)
2. Office365 integration with iX Hello is to be done at Admin level. Login to iX Hello with Organization Admin access at <https://bots.ixhello.com/>.
3. On the Administration menu, select Connection Admin. On the resulting page, to set up for Office 365, click on Admin Settings.

<figure><img src="/files/EG4gjVQmtN7dzp901Da5" alt=""><figcaption></figcaption></figure>

1. Fill the following credential details from Step and click on Save.

<figure><img src="/files/CFGgbL8nraxJDEMyabhA" alt=""><figcaption></figcaption></figure>

1. After the connection info is saved, now the employees within that organization can also use the same integration for connection with MicrosoftOffice365.
2. Click on Go to Connect System. You will be redirected to the Connect System page. Click on Add New Connection. Other employees can access the Connect System menu from Integration menu.

<figure><img src="/files/NTfsmu8rI4sEZQZuFOjr" alt="" width="563"><figcaption></figcaption></figure>

<figure><img src="/files/yniB2ncoFydb6gQ7iDbi" alt=""><figcaption></figcaption></figure>

1. Click on Add New Connection and fill out following details. Select the System Type depending upon the MicrosoftOffice365 service you are trying to connect. For example, if you are trying to connect to SharePoint, select Microsoft Office365 SharePoint.

<figure><img src="/files/5glMrTyMfnQZSLOpX0gY" alt=""><figcaption></figcaption></figure>

Once you click on “**CONNECT SYSTEM**”, you will be redirected to “**Microsoft Office365 Instance Configuration**” page as shown below:

<figure><img src="/files/M1yxJYXKl9zPMo98CEbs" alt=""><figcaption></figcaption></figure>

If you do not have this information ready, please refer to the prerequisites section.

1. The table below explains each of the inputs shown on this screen.

<table data-header-hidden><thead><tr><th width="385">Input</th><th>Information</th></tr></thead><tbody><tr><td>Input</td><td>Information</td></tr><tr><td></td><td></td></tr><tr><td>Service Instance Name</td><td>You can give any value of your choice.</td></tr><tr><td>Client ID</td><td>This is your Office 365 Application ID.</td></tr><tr><td>Client Secret</td><td>This is your application secret value.</td></tr><tr><td>Tenant Id</td><td>This is the directory ID in Office 365.</td></tr></tbody></table>

1. Fill the required information and click “**GO TO CONNECT SYSTEM**”. This will validate the details and if it is correct, integration will be saved, and user will be redirected to the connection page as shown below:

<figure><img src="/files/OdXnzEnROIWTelJdvTxm" alt=""><figcaption></figcaption></figure>

1. You will be redirected to the Microsoft login screen. Enter your Microsoft Office 365 login credentials.

![](/files/VWQ22OZGy727a0cYLj5h)

1. Microsoft Office 365 will confirm your acceptance to allow iX Hello connect to your Microsoft Office&#x20;
2. 365 Platform. Click “Allow”.
3. You will be redirected to the iX Hello integration confirmation page
4. You have successfully configured iX Hello integration to Microsoft Office 365. You can now proceed to test some of the ITSM Commands.

<figure><img src="/files/OxcNtd6Ywz4bbE4bPBEq" alt=""><figcaption></figcaption></figure>

Depending upon which service you want to access in Office365, you might need to make additions/ deletions to the API Permissions of the app. One example for Office 365 Exchange is shown in the [next document](/ixhc/general/enterprise-connections/microsoft-office-365/office-365-exchange).


# Office 365 Exchange

To access Office 365 Exchange managed APIs, you have to add permissions for the Azure registered app that we created on[ this document](/ixhc/general/enterprise-connections/microsoft-office-365).

In Azure, For Office 365 exchange permissions, you will need to have following delegated permissions:&#x20;

a. Calendars.ReadWrite.Shared,

b. People.Read,

c. Mail.ReadWrite

d. Notes.ReadWrite.All,

e. Notes.Create

f. offline\_access

&#x20;To give these permissions, click on “**Add a permission**” button, then click on “**Microsoft Graph**” button option at top and finally click on “Delegated permissions” button option and you will be able to see the list of permissions that you can give to your registered application. Find all the permissions stated above and check them and finally click on “Add permissions” button to complete the permission as shown in picture below:

![](/files/QtnHsQw48d53bRTSJcGC)

After you've delegated the above permissions for your app.


# Edit Custom Apps


# Data Sources

Data Sources allows Custom Apps to be integrated with external system connection.&#x20;

For any external system to be used as a data source, it must be first configured to integrate with VoiceWorx. You can add supported external system connection from Integration -> Connect Systems.

To add data source, select a custom app and go to Data Sources tab.

<figure><img src="/files/CRU63sumrPJWyQcZMFbX" alt="" width="298"><figcaption></figcaption></figure>

<figure><img src="/files/sryIw6UyWNmPlgIVqj8w" alt=""><figcaption></figcaption></figure>

## Components

### a. Data Source List

It features the list of already set up data sources for the app. You can perform edit/ delete operations of the data source from the action menu.

<figure><img src="/files/3KecPy7GeF8IWWs92UIl" alt=""><figcaption></figcaption></figure>

### b. Create New Data Source

To add a new data source, Click on Create New Data Source button.

<figure><img src="/files/hShjOexndQTq15hOgYEN" alt=""><figcaption></figcaption></figure>

Provide the following details:

a. Name : Name of the data source

b. Integration Configuration : The external system configuration that integrates third party to VoiceWorx.&#x20;

c. Personalized Data Source: It indicates if you want to create this data source as a personalized data source, meaning that only you can use it.  One instance where you might want to restrict usage to integration in instances where access to any other users to the integration can be an issue in data integrity etc.

<figure><img src="/files/KU8KylkofUDBhZECeLD4" alt="" width="479"><figcaption></figcaption></figure>

### c. Manage Connections

If you need to add a new integration, you can readily add the integration by clicking on Manage Connections.&#x20;

<figure><img src="/files/79WZJl5nawpYVA7VJO71" alt="" width="479"><figcaption></figcaption></figure>

<figure><img src="/files/VUhSzcFXXnqmREBrFHMY" alt=""><figcaption></figcaption></figure>

Once you've configured the required integration, you can go back to Data Source tab to add the new integration as data source for the app.


# Methods

Our next step would be to setup methods. Methods basically is constructing how the app is going to handle the user requests.&#x20;

Depending upon the type of the data source , e.g. Gen AI, Twilio data source, internal content etc., there are different operations that can be carried out.&#x20;

For instance, with Twilio Data source, the user might be able to make calls, send SMS etc., but with Gen AI data source, the user might be able to get response to queries in chat.&#x20;

Such different operations are termed as 'Request Types'.&#x20;


# Data Source : Generative AI

ixHello backs the custom apps with power of AI by integrating leading players of Gen AI into its app building mechanism. You can choose among or use the combination of available integrations with&#x20;

a. Open AI

b. Anthropic

c. Azure Open AI

d. Pinecone


# Open AI


# Text Completion

## Method Inputs

| Tab           | Parameter                      | Value                                                                                                                                                                                                                     |
| ------------- | ------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Method Inputs | Method Name                    | OpenAITextCompletionMethod                                                                                                                                                                                                |
|               | Request Type                   | openai-textcompletion                                                                                                                                                                                                     |
|               | Parameters: model              |                                                                                                                                                                                                                           |
|               | Parameters: temperature        | Controls randomness: Lowering results in less random completions. As the temperature approaches zero, the model will become deterministic and repetitive. Temperature value range between 0 to 1                          |
|               | Parameters: maxtoken           | The maximum number of tokens to generate. Requests can use up to 2,048 or 4,000 tokens shared between prompt and completion. The exact limit varies by model. (One token is roughly 4 characters for normal English text) |
|               | Parameters: top\_p             | Controls diversity via nucleus sampling: 0.5 means half of all likelihood-weighted options are considered. Top\_p value range between 0 to 1                                                                              |
|               | Parameters: frequency\_penalty | How much to penalize new tokens based on their existing frequency in the text so far.  Decreases the model's likelihood to repeat the same line verbatim. Value range between 0 2.                                        |
|               | Parameters: presence\_penalty  | How much to penalize new tokens based on whether they appear in the text so far. Increases the model's likelihood to talk about new topics. Value range between 0 2.                                                      |
|               | Parameters: question           | Enter a content that you want to generate                                                                                                                                                                                 |
|               | Parameters: OrganizationID     | {{Context.OrganizationID}}                                                                                                                                                                                                |
|               | Parameters: EmployeeID         | {{Context.EmployeeID}}                                                                                                                                                                                                    |


# Generate Images

## Method Inputs

| Tab           | Parameter                  | Value                                                |
| ------------- | -------------------------- | ---------------------------------------------------- |
| Method Inputs | Method Name                | GenerateImagesMethod                                 |
|               | Request Type               | openai-generateimages                                |
|               | Parameters: search         | any thing that you would like to generate images for |
|               | Parameters: noofimage      | Number of images you'd like to generate              |
|               | Parameters: OrganizationID | {{Context.OrganizationID}}                           |
|               | Parameters: EmployeeID     | {{Context.EmployeeID}}                               |

<figure><img src="/files/V0epM1SWsThIJv1uk9Dh" alt=""><figcaption></figcaption></figure>

## Method Output

<figure><img src="/files/hf2BylHpTMZ7lJpGsPdx" alt=""><figcaption></figcaption></figure>

## Method Result

The method result section is used to define how you want to display the result of the query response.&#x20;

| Tab           | Parameter                              | Value                                |
| ------------- | -------------------------------------- | ------------------------------------ |
| Method Result | Record JsonPath                        | $.choices\[0].message                |
|               | Valid Result Template                  | {{Records}} How else can I help you? |
|               | Record Template                        | {{content}}                          |
|               | Empty Result Template                  | How else can I help you?             |
|               | Automatically prompt for next question | False                                |

a. Record JsonPath: The result of each method is returned in JSON and ixHello uses[ JSONPath query ](https://jsonpath.com/)to access those results. By providing the value $.choices\[0].message, you are asking the method to read the message of first item in the list of results. The response JSON looks somewhat like this.

b. Valid Result Template: It defines how the result is presented to the user.&#x20;

By using {{Records}} , you are accessing all the results of the method query.&#x20;

c. Record Template : This represents a single record and how a record shall be presented to the user. {{content}}

d. Empty Result Template : In case the query does not return any value, this prompt is next displayed to the user.

e. Automatically prompt for next question: This flag is intended for integrations to smart bots like Alexa and Lex, where the user defines if the bot shall prompt the user for next question after the previous query has been answered for the app. For Generative AI apps, this flag is set to false.

<mark style="color:red;">Note:</mark> Everything parameter that has{{ }} means that the platform is interpreting the value from its parameters, input parameters or managed results.

<figure><img src="/files/KlyZxcGUgZM9jDEVJZil" alt=""><figcaption></figcaption></figure>


# Chat Completion

## Method Inputs

| Tab           | Parameter                      | Value                                                                                                                                                                                                                     |
| ------------- | ------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Method Inputs | Method Name                    | OpenAIChatCompletionMethod                                                                                                                                                                                                |
|               | Request Type                   | openai-textcompletion                                                                                                                                                                                                     |
|               | Parameters: model              | gpt-3.5-turbo-16k                                                                                                                                                                                                         |
|               | Parameters: temperature        | Controls randomness: Lowering results in less random completions. As the temperature approaches zero, the model will become deterministic and repetitive. Temperature value range between 0 to 1                          |
|               | Parameters: maxtoken           | The maximum number of tokens to generate. Requests can use up to 2,048 or 4,000 tokens shared between prompt and completion. The exact limit varies by model. (One token is roughly 4 characters for normal English text) |
|               | Parameters: top\_p             | Controls diversity via nucleus sampling: 0.5 means half of all likelihood-weighted options are considered. Top\_p value range between 0 to 1                                                                              |
|               | Parameters: frequency\_penalty | How much to penalize new tokens based on their existing frequency in the text so far.  Decreases the model's likelihood to repeat the same line verbatim. Value range between 0 2.                                        |
|               | Parameters: presence\_penalty  | How much to penalize new tokens based on whether they appear in the text so far. Increases the model's likelihood to talk about new topics. Value range between 0 2.                                                      |
|               | Parameters: systemMessage      | Provides a guardrail for the app                                                                                                                                                                                          |
|               | Parameters: userQuestion       | ask any question                                                                                                                                                                                                          |
|               | Parameters: sessionTime        | Session time in minutes used to maintain conversation history. e.g. 10                                                                                                                                                    |
|               | Parameters: OrganizationID     | {{Context.OrganizationID}}                                                                                                                                                                                                |
|               | Parameters: EmployeeID         | {{Context.EmployeeID}}                                                                                                                                                                                                    |


# Create Embedding

## Method Inputs

| Tab           | Parameter                  | Value                               |
| ------------- | -------------------------- | ----------------------------------- |
| Method Inputs | Method Name                | OpenAICreateEmbedding               |
|               | Request Type               | openai-createembedding              |
|               | Parameters: model          | text-embedding-ada-002              |
|               | Parameters: input          | Enter a content to create embedding |
|               | Parameters: OrganizationID | {{Context.OrganizationID}}          |
|               | Parameters: EmployeeID     | {{Context.EmployeeID}}              |


# OpenAI Assistant

## Method Input

| Tab           | Parameter                  | Value                                                                                                                                                                                            |
| ------------- | -------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| Method Inputs | Method Name                | OpenAIAssistantMethod                                                                                                                                                                            |
|               | Request Type               | openai-assistant                                                                                                                                                                                 |
|               | Parameters: model          | A suitable model of gpt as per the app requirement                                                                                                                                               |
|               | Parameters: systemMessage  | This defines the guardrail for the app. In the example below, it's taking from a internal content JSON file and managed parameters.{{Session.StoreDescription}} ,{{App.AIContent.SystemMessage}} |
|               | Parameters: userQuestion   | The input query from the user. It could also be plain text or it could be derived from input parameters.e.g. 'what services does ACME provide?' or {{UserInput}                                  |
|               | Parameters: streaming      | False                                                                                                                                                                                            |
|               | Parameters: sessionTime    | 10 ( Time inminutes which will be used to maintain conversation history)                                                                                                                         |
|               | Parameters: OrganizationID | {{Context.OrganizationID}}                                                                                                                                                                       |
|               | Parameters: EmployeeID     | {{Context.EmployeeID}}                                                                                                                                                                           |

<figure><img src="/files/mWVl4yhxJTJ5BUTaauOe" alt=""><figcaption></figcaption></figure>


# Anthropic


# Chat Completion

## Method Inputs

| Tab           | Parameter                      | Value                                                                                                                                                                                                                     |
| ------------- | ------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Method Inputs | Method Name                    | OpenAIChatCompletionMethod                                                                                                                                                                                                |
|               | Request Type                   | openai-textcompletion                                                                                                                                                                                                     |
|               | Parameters: model              | gpt-3.5-turbo-16k                                                                                                                                                                                                         |
|               | Parameters: temperature        | Controls randomness: Lowering results in less random completions. As the temperature approaches zero, the model will become deterministic and repetitive. Temperature value range between 0 to 1                          |
|               | Parameters: maxtoken           | The maximum number of tokens to generate. Requests can use up to 2,048 or 4,000 tokens shared between prompt and completion. The exact limit varies by model. (One token is roughly 4 characters for normal English text) |
|               | Parameters: top\_p             | Controls diversity via nucleus sampling: 0.5 means half of all likelihood-weighted options are considered. Top\_p value range between 0 to 1                                                                              |
|               | Parameters: frequency\_penalty | How much to penalize new tokens based on their existing frequency in the text so far.  Decreases the model's likelihood to repeat the same line verbatim. Value range between 0 2.                                        |
|               | Parameters: presence\_penalty  | How much to penalize new tokens based on whether they appear in the text so far. Increases the model's likelihood to talk about new topics. Value range between 0 2.                                                      |
|               | Parameters: systemMessage      | Provides a guardrail for the app                                                                                                                                                                                          |
|               | Parameters: userQuestion       | ask any question                                                                                                                                                                                                          |
|               | Parameters: sessionTime        | Session time in minutes used to maintain conversation history. e.g. 10                                                                                                                                                    |
|               | Parameters: OrganizationID     | {{Context.OrganizationID}}                                                                                                                                                                                                |
|               | Parameters: EmployeeID         | {{Context.EmployeeID}}                                                                                                                                                                                                    |


# Azure Open AI


# Chat Completion

## Method Inputs

| Tab          | Parameter                  | Value                                                                  |
| ------------ | -------------------------- | ---------------------------------------------------------------------- |
| Method Input | Method Name                | AzureChatCompletionMethod                                              |
|              | Request Type               | azure-openai-chatcompletion                                            |
|              | Parameters: SystemMessage  | {{Sesssion.StoreDescription}}, {{App.AIContent.SystemMessage}}         |
|              | Parameters: UserQuestion   | {{UserInput}}                                                          |
|              | Parameters: Streaming      | False                                                                  |
|              | Parameters: SessionTime    | Session time in minutes used to maintain conversation history. e.g. 10 |
|              | Parameters: OrganizationID | {{Context.OrganizationID}}                                             |
|              | Parameters: EmployeeID     | {{Context.EmployeeID}}                                                 |

<figure><img src="/files/uHMXIxCYcsEI3EV8kWzn" alt=""><figcaption></figcaption></figure>

## Method Output

You can get the result of the query as a single record or you can get a list of records. You can also define the number of items in the list using Record Limit Parameter.

<figure><img src="/files/IxQuLUQOMGu6TNfhjGbI" alt=""><figcaption></figcaption></figure>

## Method Result

The method result section is used to define how you want to display the result of the query response.&#x20;

| Tab           | Parameter                              | Value                                |
| ------------- | -------------------------------------- | ------------------------------------ |
| Method Result | Record JsonPath                        | $.choices\[0].message                |
|               | Valid Result Template                  | {{Records}} How else can I help you? |
|               | Record Template                        | {{content}}                          |
|               | Empty Result Template                  | How else can I help you?             |
|               | Automatically prompt for next question | False                                |

a. Record JsonPath: The result of each method is returned in JSON and ixHello uses[ JSONPath query ](https://jsonpath.com/)to access those results. By providing the value $.choices\[0].message, you are asking the method to read the message of first item in the list of results. The response JSON looks somewhat like this.

b. Valid Result Template: It defines how the result is presented to the user.&#x20;

By using {{Records}} , you are accessing all the results of the method query.&#x20;

c. Record Template : This represents a single record and how a record shall be presented to the user. {{content}}

d. Empty Result Template : In case the query does not return any value, this prompt is next displayed to the user.

e. Automatically prompt for next question: This flag is intended for integrations to smart bots like Alexa and Lex, where the user defines if the bot shall prompt the user for next question after the previous query has been answered for the app. For Generative AI apps, this flag is set to false.

<mark style="color:red;">Note:</mark> Everything parameter that has{{ }} means that the platform is interpreting the value from its parameters, input parameters or managed results.

<figure><img src="/files/yWrQmDCRYavuB0kEkCBY" alt=""><figcaption></figcaption></figure>


# Chat Completion RAG

For this setup, you have to take care of following things

## Method Input

| Tab           | Parameter                  | Value                                                                                                                                                                                            |
| ------------- | -------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| Method Inputs | Method Name                | AzureChatRAGMethod                                                                                                                                                                               |
|               | Request Type               | azure-chatcompletion-rag                                                                                                                                                                         |
|               | Parameters: model          | A suitable GPT model as per your use case, e.g. gpt-3.5-turbo-16k                                                                                                                                |
|               | Parameters: systemMessage  | This defines the guardrail for the app. In the example below, it's taking from a internal content JSON file and managed parameters.{{Session.StoreDescription}} ,{{App.AIContent.SystemMessage}} |
|               | Parameters :userQuestion   | The input query from the user. It could also be plain text or it could be derived from input parameters.e.g. 'what services does ACME provide?' or {{UserInput}}                                 |
|               | Parameters: streaming      | False                                                                                                                                                                                            |
|               | Parameters: sessionTime    | 10 ( Time inminutes which will be used to maintain conversation history)                                                                                                                         |
|               | Parameters: OrganizationID | {{Context.OrganizationID}}                                                                                                                                                                       |
|               | Parameters: EmployeeID     | {{Context.EmployeeID}}                                                                                                                                                                           |

<figure><img src="/files/Qt9ogI6pLTb4nO6rPgwR" alt=""><figcaption></figcaption></figure>

## Method Output

You can get the result of the query as a single record or you can get a list of records. You can also define the number of items in the list using Record Limit Parameter.

<figure><img src="/files/laNEYknA9uL6WsiMIlw0" alt=""><figcaption></figcaption></figure>

## Method Result

The method result section is used to define how you want to display the result of the query response.&#x20;

| Tab           | Parameter                              | Value                                |
| ------------- | -------------------------------------- | ------------------------------------ |
| Method Result | Record JsonPath                        | $.choices\[0].message                |
|               | Valid Result Template                  | {{Records}} How else can I help you? |
|               | Record Template                        | {{content}}                          |
|               | Empty Result Template                  | How else can I help you?             |
|               | Automatically prompt for next question | False                                |

a. Record JsonPath: The result of each method is returned in JSON and ixHello uses[ JSONPath query ](https://jsonpath.com/)to access those results. By providing the value $.choices\[0].message, you are asking the method to read the message of first item in the list of results. The response JSON looks somewhat like this.

b. Valid Result Template: It defines how the result is presented to the user.&#x20;

By using {{Records}} , you are accessing all the results of the method query.&#x20;

c. Record Template : This represents a single record and how a record shall be presented to the user. {{content}}

d. Empty Result Template : In case the query does not return any value, this prompt is next displayed to the user.

e. Automatically prompt for next question: This flag is intended for integrations to smart bots like Alexa and Lex, where the user defines if the bot shall prompt the user for next question after the previous query has been answered for the app. For Generative AI apps, this flag is set to false.

<mark style="color:red;">Note:</mark> Everything parameter that has{{ }} means that the platform is interpreting the value from its parameters, input parameters or managed results.

<figure><img src="/files/CBL1PkyTMqZirft8GW8N" alt=""><figcaption></figcaption></figure>


# Azure OpenAI Assistant

## Method Inputs

| Tab           | Parameter                  | Value                                |
| ------------- | -------------------------- | ------------------------------------ |
| Method Inputs | Method Name                | ChatCompletionMethod                 |
|               | Request Type               | azure-openai-assistant               |
|               | Parameters: model          | gpt-35-turbo-16k                     |
|               | Parameters: systemMessage  | App guardrails shall be derived from |
|               | Parameters: userQuestion   | Question from user                   |
|               | Parameters: streaming      | False                                |
|               | Parameters: sessionTime    |                                      |
|               | Parameters: OrganizationID | {{Context.OrganizationID}}           |
|               | Parameters: EmployeeID     | {{Context.EmployeeID}}               |

<figure><img src="/files/UwqRS7qTiBjLfKcPD7YD" alt=""><figcaption></figcaption></figure>

## Method Output

<figure><img src="/files/5C0B2Z0c0PeF4vGlk61H" alt=""><figcaption></figcaption></figure>

## Method Result

| Tab           | Parameter                              | Value                       |
| ------------- | -------------------------------------- | --------------------------- |
| Method Result | Record JsonPath                        | $.data\[0].content\[0].text |
|               | Valid Result Template                  | {{Records}}                 |
|               | Record Template                        | {{value}}                   |
|               | Empty Result Template                  | How else can I help you?    |
|               | Automatically prompt for next question | False                       |

<figure><img src="/files/g47wJvZ4PPRNCUhGqWSK" alt=""><figcaption></figcaption></figure>


# Pinecone


# Upsert

## Method Inputs

| Tab           | Parameter                  | Value                      |
| ------------- | -------------------------- | -------------------------- |
| Method Inputs | Method Name                | PineconeUpsert             |
|               | Request Type               | pinecone-upsert            |
|               | Parameters: ids            |                            |
|               | Parameters: title          |                            |
|               | Parameters: input          |                            |
|               | Parameters: embeddedData   |                            |
|               | Parameters: OrganizationID | {{Context.OrganizationID}} |
|               | Parameters: EmployeeID     | {{Context.EmployeeID}}     |


# Query

## Method Inputs

| Tab           | Parameter                  | Value                      |
| ------------- | -------------------------- | -------------------------- |
| Method Inputs | Method Name                | PineconeQueryMethod        |
|               | Request Type               | pinecone-query             |
|               | Parameters: embeddedData   |                            |
|               | Parameters: OrganizationID | {{Context.OrganizationID}} |
|               | Parameters: EmployeeID     | {{Context.EmployeeID}}     |


# Train from HTML page

## Method Inputs

| Tab           | Parameter                  | Value                                                                                                                                     |
| ------------- | -------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------- |
| Method Inputs | Method Name                | PineconeTrainfromHtmlPage                                                                                                                 |
|               | Request Type               | pinecone-trainfromhtmlpage                                                                                                                |
|               | Parameters: url            | Webpage URL                                                                                                                               |
|               | Parameters: chunkSize      | Number of words in a chunk                                                                                                                |
|               | Parameters: chunkOverlap   | Number of words in a chunk that are allowed to overlap. It is used to make sure that the context is not missed when saving data as chunks |
|               | Parameters: OrganizationID | {{Context.OrganizationID}}                                                                                                                |
|               | Parameters : EmployeeID    | {{Context.EmployeeID}}                                                                                                                    |


# Data Source : Salesforce

Salesforce objects and fields are analogous to database tables and the table columns.&#x20;

Objects already created by Salesforce are called standard objects.

Objects that you create in your organization with the user interface or with the Metadata API are called custom objects.

Objects you create that map to data stored outside your organization are called external objects.

You can access the Objects and Fields in your salesforce account, so use them as data source for your custom app.

## Prerequisites

a. Salesforce Account

b. Integrate ixHello to Salesforce Account

c. Create Custom App

d. Create Data Source

## Request Types

{% tabs %}
{% tab title="Salesforce" %}
a. [GetAllObjects](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/getallobjects)

b. [GetObjectFields](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/getobjectfields)

c.[GetObjectById](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/getobjectbyid)

d.[GetObjectAllRecords](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/getobjectallrecords)

e. [GetSOQLQueryResult](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/getsoqlqueryresult)

f. [CreateObject](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/createobject)

g.[UpdateObject](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/updateobject)

h.[DeleteObject](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/deleteobject)
{% endtab %}

{% tab title="Salesforce Einstein" %}
a. [ExecutSAQLQuery](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/einstein.executesaqlquery)

b. [ExecuteLens](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/einstein.executelens)

c. [GetAnalyticsResources](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/einstein.getanalyticsresources)

d. [GetDashboards](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/einstein.getdashboards)

e. [GetDataSets](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/einstein.getdatasets)

f. [GetLenses](/ixhc/general/edit-custom-apps/methods/data-source-salesforce/einstein.getlenses)
{% endtab %}

{% tab title="Salesforce ServiceCloud" %}

{% endtab %}

{% tab title="Tableau" %}

{% endtab %}
{% endtabs %}


# GetAllObjects

This method returns all the objects you have in your salesforce account and it does not require any parameter.

**Parameters:**

| Tab           | Parameter    | Value               |
| ------------- | ------------ | ------------------- |
| Method Inputs | Method Name  | GetAllObjectsMethod |
|               | Request Type | GetAllObjects       |
|               | Parameters   | N/A                 |

<figure><img src="/files/fAa08xz8NeQruHEaHrsL" alt=""><figcaption></figcaption></figure>

#### Sample Response:

```json
// Sample Response
[
 {
 "activateable": false,
 "createable": false,
 "custom": false,
 "customSetting": false,
 "deepCloneable": false,
 "deletable": false,
 "deprecatedAndHidden": false,
 "feedEnabled": false,
 "hasSubtypes": false,
 "isInterface": false,
 "isSubtype": false,
 "keyPrefix": null,
 "label": "Accepted Event Relation",
 "labelPlural": "Accepted Event Relations",
 "layoutable": false,
 "mergeable": false,
 "mruEnabled": false,
 "name": "AcceptedEventRelation",
 "queryable": true,
 "replicateable": false,
 "retrieveable": true,
 "searchable": false,
 "triggerable": false,
 "undeletable": false,
 "updateable": false,
 "urls": {
 "rowTemplate": "/services/data/v47.0/sobjects/AcceptedEventRelation/{ID}",
 "defaultValues": "/services/data/v47.0/sobjects/AcceptedEventRelation/defaultValues?recordTypeId&fields",
 "describe": "/services/data/v47.0/sobjects/AcceptedEventRelation/describe",
 "sobject": "/services/data/v47.0/sobjects/AcceptedEventRelation"
 }
 },
{another record},
{another record},
]
```


# GetObjectFields

This method is used to fetch all the available fields for the specific Salesforce Object.

#### Parameters:

| Method Inputs | Method Name            | GetObjectFieldsMethod                             |
| ------------- | ---------------------- | ------------------------------------------------- |
|               | Request Type           | GetObjectFields                                   |
|               | Parameters: ObjectName | <p>Salesforce Object name </p><p>e.g. Contact</p> |

<figure><img src="/files/drR4beBLOij8QRb0J7xk" alt=""><figcaption></figcaption></figure>

#### Sample Response:

```json
// Sample Response
[
{
 "aggregatable": true,
 "aiPredictionField": false,
 "autoNumber": false,
 "byteLength": 120,
 "calculated": false,
 "calculatedFormula": null,
 "cascadeDelete": false,
 "caseSensitive": false,
 "compoundFieldName": "Name",
 "controllerName": null,
 "createable": true,
 "custom": false,
 "defaultValue": null,
 "defaultValueFormula": null,
 "defaultedOnCreate": false,
 "dependentPicklist": false,
 "deprecatedAndHidden": false,
 "digits": 0,
 "displayLocationInDecimal": false,
 "encrypted": false,
 "externalId": false,
 "extraTypeInfo": "personname",
 "filterable": true,
 "filteredLookupInfo": null,
 "formulaTreatNullNumberAsZero": false,
 "groupable": true,
 "highScaleNumber": false,
 "htmlFormatted": false,
 "idLookup": false,
 "inlineHelpText": null,
 "label": "First Name",
 "length": 40,
 "mask": null,
 "maskType": null,
 "name": "FirstName",
 "nameField": false,
 "namePointing": false,
 "nillable": true,
 "permissionable": false,
 "picklistValues": [],
 "polymorphicForeignKey": false,
 "precision": 0,
 "queryByDistance": false,
 "referenceTargetField": null,
 "referenceTo": [],
 "relationshipName": null,
 "relationshipOrder": null,
 "restrictedDelete": false,
 "restrictedPicklist": false,
 "scale": 0,
 "searchPrefilterable": false,
 "soapType": "xsd:string",
 "sortable": true,
 "type": "string",
 "unique": false,
 "updateable": true,
 "writeRequiresMasterRead": false
 },
{record for another field},
{record for another field},…
]
```

Contact object has approximately 61 fields. This method will return those records, one record represents one field.


# GetObjectById

This method is used to fetch one specific record for the specific Salesforce Object like Contact or Opportunity

#### Parameters:

| MethodInputs | Method Name            | GetObjectById                                                                         |
| ------------ | ---------------------- | ------------------------------------------------------------------------------------- |
|              | Request Type           | GetObjectById                                                                         |
|              | Parameters: ObjectName | <p>Salesforce Object name</p><p>e.g. Contact</p>                                      |
|              | Parameters: Id         | <p>Salesforce Object Id</p><p>e.g. 032w000002GFOPAA4</p>                              |
|              | Parameters: FieldNames | <p>One or more fields separated by comma </p><p>e.g.Id, FirstName, LastName,Phone</p> |

#### Sample Response:

```json
// Sample Response
[
 {
 "Id": "0032w000002GFOPAA4",
 "FirstName": "Rose",
 "LastName": "Gonzalez",
 "Phone": "(512) 757-6000"
 }
]
```


# GetObjectAllRecords

This method is used to fetch all the records for the specific Salesforce Object like Contact or Opportunity.

#### Parameters:

| Method Inputs | Method Name             | GetObjectAllRecords                                                                    |
| ------------- | ----------------------- | -------------------------------------------------------------------------------------- |
|               | Request Type            | GetObjectAllRecords                                                                    |
|               | Parameters : ObjectName | <p>Salesforce Object name</p><p>e.g. Contact</p>                                       |
|               | Parameters: FieldNames  | <p>One or more fields separated by comma </p><p>e.g. Id, FirstName, LastName,Phone</p> |

#### Sample Response:

```json
// Sample Response
[
 {
 "attributes": {
 "type": "Contact",
 "url": "/services/data/v47.0/sobjects/Contact/0032w000002GFOPAA4"
 },
 "Id": "0032w000002GFOPAA4",
 "FirstName": "Rose",
 "LastName": "Gonzalez",
 "Phone": "(512) 757-6000"
 },
 {
 "attributes": {
 "type": "Contact",
 "url": "/services/data/v47.0/sobjects/Contact/0032w000002GFOQAA4"
 },
 "Id": "0032w000002GFOQAA4",
 "FirstName": "Sean",
 "LastName": "Forbes",
 "Phone": "(512) 757-6000"
 }
]
```


# GetSOQLQueryResult

GetSOQLQueryResult request type allows you to run Standard Salesforce Object Search Language (SOQL) query against your salesforce objects and fields. It is suitable for complex operations and is basically running your custom queries, getting desired results and then using them as input for your app.

## Method Inputs

| Tab           | Parameter             | Value                                                                                                                                                                                                                                   |
| ------------- | --------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Method Inputs | Method Name           | <p>You can give any name. This name will be displayed in the methods list.</p><p>e.g.GetSOQLQueryResult</p>                                                                                                                             |
|               | Request Type          | GetSOQLQueryResult                                                                                                                                                                                                                      |
|               | Parameters: SQQLQuery | <p>This is a standard Salesforce Object Search language (SOQL) query which you can use in different ways </p><p>e.g. SELECT Id, FirstName, LastName,Phone</p><p>FROM Contact</p><p>WHERE FirstName='John'</p><p>and LastName='Bond'</p> |

## Method Output

| Parameter    | Info                                                                                                                                          |
| ------------ | --------------------------------------------------------------------------------------------------------------------------------------------- |
| Result Type  | <p>ListOfRecords (Auto Selected)</p><p>It indicates that the method can return one or more records</p>                                        |
| Record Limit | <p>Numeric value (e.g. 10)</p><p>If Method is returning larger number of records, then you can limit this result by providing this value.</p> |

#### Sample Response:

```json
// Sample Response
[
 {
 "attributes": {
 "type": "Contact",
 "url": "/services/data/v47.0/sobjects/Contact/0032w000002GFOVAA4"
 },
 "Id": "0032w000002GFOVAA4",
 "FirstName": "John",
 "LastName": "Bond",
 "Phone": "(312) 596-1000"
 }
]
```

## Method Result

| Tab | Parameter             | Value                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             |
| --- | --------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|     | Record JsonPath       | <p>Each method returns data in json format and to pick any specific information from the json, we need to specify the <em>Record JsonPath</em> so Voice Worx platform will pick that data in consideration while displaying results.</p><p>Example: $.\[\*]</p><p>This selects all the records that is returned</p>                                                                                                                                                                                                                                                                                                                                                                                                               |
|     | Empty Result Template | <p>When the method does not return any data, this response will be generated. So we can specify something like:</p><p>“Sorry, I could not find the contact details”</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           |
|     | Record Template       | <p>"type": "Contact",</p><p> "url": "/services/data/v47.0/sobjects/Contact/0032w000002GFOVAA4"</p><p> },</p><p> "Id": "0032w000002GFOVAA4",</p><p> "FirstName": "John",</p><p> "LastName": "Bond",</p><p> "Phone": "(312) 596-1000"</p><p> }</p><p>In this example, Phone is at root level and can be accessed by {{Phone}}</p><p>So record template can be written as</p><p><em>Phone number of {{FirstName}} {{LastName}} is {{Phone}}</em></p><p>So, when a method is executed, this will be displayed as,</p><p><em>Phone number of John Bond is (312) 596-1000</em></p><p>Note:</p><p>Anything written in between {{ }} is represented as a json field and it’s case-sensitive.</p><p>{{Phone}} is not same as {{phone}}</p> |
|     | Result Template       | <p>This indicates how method results should be represented.</p><p>Here you go, {{Records}}</p><p>Here, {{Records}} represents one or more records separated by a separator.</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |


# CreateObject

This method is used to create a new record for the Salesforce Object like Contact or Opportunity

#### Parameters:

## Method Inputs

| Method Inputs | Method Name             | CreateObjectMethod                                                                                                                                      |
| ------------- | ----------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- |
|               | Request Type            | CreateObject                                                                                                                                            |
|               | Parameters: ObjectName  | <p>Salesforce Object name </p><p>e.g. Contact</p>                                                                                                       |
|               | Paarameters: RecordJson | <p>A valid json with valid field names of the Salesforce Object. </p><p>e.g. {"FirstName": "John", "LastName": "Green", "Phone": "(312) 596-2000" }</p> |

#### Sample Response:

```json
// Sample Response
{
 "id": "0032w00000FIZCxAAP",
 "success": true,
 "errors": []
}
```


# UpdateObject

This method is used to create a new record for the Salesforce Object like Contact or Opportunity

## Method Inputs

| Method Input | Method Name           | UpdateObject                                                                                                                                            |
| ------------ | --------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- |
|              | Request Type          | UpdateObject                                                                                                                                            |
|              | Parameter: ObjectName | <p>Salesforce Object name </p><p>e.g. Contact</p>                                                                                                       |
|              | Parameter : RecordId  | <p>Salesforce Object Id </p><p>e.g. 0032w00000FIZCxAAP</p>                                                                                              |
|              | Parameter: RecordJson | <p>A valid json with valid field names of the Salesforce Object. </p><p>e.g. {"FirstName": "John", "LastName": "Green", "Phone": "(312) 596-2000" }</p> |

#### Sample Response:

```json
// Sample Response
{
 "id": "",
 "success": true,
 "errors": ""
}
```


# DeleteObject

This method is used to Delete any specific record of Salesforce Object like Contact

## Method Inputs

| Method Inputs | Method Name           | DeleteObjectMethod                                      |
| ------------- | --------------------- | ------------------------------------------------------- |
|               | Request Type          | DeleteObject                                            |
|               | Parameter: ObjectName | <p>Salesforce Object name </p><p>e.g. Contact</p>       |
|               | Parameter: RecordId   | <p>A valid record id </p><p>e.g. 0032w00000FIZCxAAP</p> |

#### Sample Response:

```json
// Some code
{
 "success": true
}
```


# Einstein.ExecuteSAQLQuery

| Parameter             | Info                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         |
| --------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|                       |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              |
| Method Name           | You can give any name. This name will be displayed in the methods list.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      |
| Request Type          | <p><strong>Einstein.ExecuteSAQLQuery</strong></p><p>This is the method that you are using</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                |
| Parameters            | As explained in the previous table                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           |
| Result Type           | <p>ListOfRecords (Auto Selected)</p><p>It indicates that the method can return one or more records</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       |
| Record Limit          | <p>Numeric value (e.g. 10)</p><p>If Method is returning larger number of records, then you can limit this result by providing this value.</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                |
| Record JsonPath       | <p>Each method returns data in json format and to pick any specific information from the json, we need to specify the <em>Record JsonPath</em> so Voice Worx platform will pick that data in consideration while displaying results.</p><p>Example: $.results.records\[\*]</p><p>This selects all the records that is returned.</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          |
| Empty Result Template | <p>When the method does not return any data, this response will be generated. So we can specify something like:</p><p>“Sorry, I could not find the data”</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 |
| Record Template       | <p>Let’s assume that the following json represents a record.</p><p>{</p><ul><li>"action" : query,</li><li>"responseId" : 4ZYb6WZm8DGZp\_0fHwqHSV,</li><li><p>"results" : -{</p><ul><li>"metadata" : +\[ ... ],</li><li><p>"records" : -\[</p><ul><li><p>-{</p><ul><li>"Quarter" : This Quarter,</li><li>"StageName" : Prospecting,</li><li>"stage\_sort" : 1,</li><li>"sum\_Amount" : 24597662,</li><li>"time\_sort" : 1</li></ul></li></ul></li></ul></li></ul><p>},</p><ul><li><p></p><ul><li><p></p><ul><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... }</li></ul></li></ul></li></ul><p>]</p><ul><li>},</li><li>"query" : q = load "0Fb2w000000HGNMCA4/0Fc2w000002C0f9CAC";\nq = filter q by 'Stage.IsClosed' == "false";\nq = filter q by 'ForecastCategory' != "Omitted";\ntoday = filter q by 'IsLastUpdate' == "1";\ntoday = group today by 'StageName';\ntoday = foreach today generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 0 as 'time\_sort', "Current" as 'Quarter';\none = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."1 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["current quarter"..]);\none = group one by 'StageName';\none = foreach one generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 1 as 'time\_sort', "This Quarter" as 'Quarter';\ntwo = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."2 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["1 quarter ago"..]);\ntwo = group two by 'StageName';\ntwo = foreach two generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 2 as 'time\_sort', "1 quarter ago" as 'Quarter';\nthree = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."3 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["2 quarters ago"..]);\nthree = group three by 'StageName';\nthree = foreach three generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 3 as 'time\_sort', "2 quarters ago" as 'Quarter';\nfour = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."4 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["3 quarters ago"..]);\nfour = group four by 'StageName';\nfour = foreach four generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 4 as 'time\_sort', "3 quarters ago" as 'Quarter';\nresult = union today, one, two, three, four;\nresult = group result by ('Quarter', 'StageName');\nresult = foreach result generate 'StageName' as 'StageName', 'Quarter' as 'Quarter', sum('sum\_Amount') as 'sum\_Amount', first('stage\_sort') as 'stage\_sort', first('time\_sort') as 'time\_sort';\nresults = order result by ('time\_sort' desc, 'stage\_sort' asc);,</li><li>"responseTime" : 114</li></ul><p>}</p><p>In this example, you want to access the items of records array,</p><p>So record template can be written as</p><p><em>{{StageName}}, amount {{sum\_Amount}}</em></p><p>Note:</p><p>Anything written in between {{ }} is represented as a json field and it’s case-sensitive.</p><p>{{StageName}} is not same as {{stagename}}</p> |
| Result Template       | <p>This indicates how method results should be represented.</p><p>Here you go, We have found the following historical pipeline by stage, {{Records}}</p><p>Here, {{Records}} represents one or more records separated by a separator.</p><p>Now the response looks like:</p><p>I have found the following historical pipeline by stage, Prospecting, amount 24597662, Qualification, amount 53942843, Needs Analysis, amount 64842060, Value Proposition, amount 29715331, Perception Analysis, amount 38981892, Proposal/Price Quote, amount 15889140, Negotiation/Review, amount 17044891, Prospecting, amount 24597662, Qualification, amount 53942843, Needs Analysis, amount 64842060</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |

###


# Einstein.ExecuteLens

This method requires parameters as explained below.

Parameters:

| Parameter | Sample Value                                             | Info                                    |
| --------- | -------------------------------------------------------- | --------------------------------------- |
|           |                                                          |                                         |
| Lens      | Historical Pipeline Stage by Quarter (VoiceworxSalesApp) | Choose the required lens from dropdown. |

Other than these two parameters, you need to provide following standard inputs

| Parameter             | Info                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         |
| --------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|                       |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                              |
| Method Name           | You can give any name. This name will be displayed in the methods list.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      |
| Request Type          | <p><strong>Einstein.ExecuteLens</strong></p><p>This is the method that you are using</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                     |
| Parameters            | As explained in the previous table                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           |
| Result Type           | <p>ListOfRecords (Auto Selected)</p><p>It indicates that the method can return one or more records</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       |
| Record Limit          | <p>Numeric value (e.g. 10)</p><p>If Method is returning larger number of records, then you can limit this result by providing this value.</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                |
| Record JsonPath       | <p>Each method returns data in json format and to pick any specific information from the json, we need to specify the <em>Record JsonPath</em> so Voice Worx platform will pick that data in consideration while displaying results.</p><p>Example: $.results.records\[\*]</p><p>This selects all the records that is returned.</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          |
| Empty Result Template | <p>When the method does not return any data, this response will be generated. So we can specify something like:</p><p>“Sorry, I could not find the data”</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 |
| Record Template       | <p>Let’s assume that the following json represents a record.</p><p>{</p><ul><li>"action" : query,</li><li>"responseId" : 4ZYb6WZm8DGZp\_0fHwqHSV,</li><li><p>"results" : -{</p><ul><li>"metadata" : +\[ ... ],</li><li><p>"records" : -\[</p><ul><li><p>-{</p><ul><li>"Quarter" : This Quarter,</li><li>"StageName" : Prospecting,</li><li>"stage\_sort" : 1,</li><li>"sum\_Amount" : 24597662,</li><li>"time\_sort" : 1</li></ul></li></ul></li></ul></li></ul><p>},</p><ul><li><p></p><ul><li><p></p><ul><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... },</li><li>+{ ... }</li></ul></li></ul></li></ul><p>]</p><ul><li>},</li><li>"query" : q = load "0Fb2w000000HGNMCA4/0Fc2w000002C0f9CAC";\nq = filter q by 'Stage.IsClosed' == "false";\nq = filter q by 'ForecastCategory' != "Omitted";\ntoday = filter q by 'IsLastUpdate' == "1";\ntoday = group today by 'StageName';\ntoday = foreach today generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 0 as 'time\_sort', "Current" as 'Quarter';\none = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."1 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["current quarter"..]);\none = group one by 'StageName';\none = foreach one generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 1 as 'time\_sort', "This Quarter" as 'Quarter';\ntwo = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."2 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["1 quarter ago"..]);\ntwo = group two by 'StageName';\ntwo = foreach two generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 2 as 'time\_sort', "1 quarter ago" as 'Quarter';\nthree = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."3 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["2 quarters ago"..]);\nthree = group three by 'StageName';\nthree = foreach three generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 3 as 'time\_sort', "2 quarters ago" as 'Quarter';\nfour = filter q by (date('ValidFromDate\_Year', 'ValidFromDate\_Month', 'ValidFromDate\_Day') in \[.."4 quarters ago"]) && (date('ValidToDate\_Year', 'ValidToDate\_Month', 'ValidToDate\_Day') in \["3 quarters ago"..]);\nfour = group four by 'StageName';\nfour = foreach four generate 'StageName' as 'StageName', sum('Amount') as 'sum\_Amount', last('Opportunity.Stage.SortOrder') as 'stage\_sort', 4 as 'time\_sort', "3 quarters ago" as 'Quarter';\nresult = union today, one, two, three, four;\nresult = group result by ('Quarter', 'StageName');\nresult = foreach result generate 'StageName' as 'StageName', 'Quarter' as 'Quarter', sum('sum\_Amount') as 'sum\_Amount', first('stage\_sort') as 'stage\_sort', first('time\_sort') as 'time\_sort';\nresults = order result by ('time\_sort' desc, 'stage\_sort' asc);,</li><li>"responseTime" : 114</li></ul><p>}</p><p>In this example, you want to access the items of records array,</p><p>So record template can be written as</p><p><em>{{StageName}}, amount {{sum\_Amount}}</em></p><p>Note:</p><p>Anything written in between {{ }} is represented as a json field and it’s case-sensitive.</p><p>{{StageName}} is not same as {{stagename}}</p> |
| Result Template       | <p>This indicates how method results should be represented.</p><p>Here you go, We have found the following historical pipeline by stage, {{Records}}</p><p>Here, {{Records}} represents one or more records separated by a separator.</p><p>Now the response looks like:</p><p>I have found the following historical pipeline by stage, Prospecting, amount 24597662, Qualification, amount 53942843, Needs Analysis, amount 64842060, Value Proposition, amount 29715331, Perception Analysis, amount 38981892, Proposal/Price Quote, amount 15889140, Negotiation/Review, amount 17044891, Prospecting, amount 24597662, Qualification, amount 53942843, Needs Analysis, amount 64842060</p>                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |

###


# Einstein.GetAnalyticsResources

This method is used to fetch all the available analytics resources like datasets,lenses,dashboards,etc.

#### Parameters:

This method does not require any parameter.

#### Sample Response:

Contact object has around 61 fields so this method will return you that much records, one record represents one field.

```json
{
 "dashboards": "/services/data/v49.0/wave/dashboards",
 "dataConnectorTypes": "/services/data/v49.0/wave/dataConnectorTypes",
 "dataConnectors": "/services/data/v49.0/wave/dataConnectors",
 "dataflowjobs": "/services/data/v49.0/wave/dataflowjobs",
 "datasets": "/services/data/v49.0/wave/datasets",
 "folders": "/services/data/v49.0/wave/folders",
 "lenses": "/services/data/v49.0/wave/lenses",
 "query": "/services/data/v49.0/wave/query",
 "recipes": "/services/data/v49.0/wave/recipes",
 "replicatedDatasets": "/services/data/v49.0/wave/replicatedDatasets",
 "schedules": "/services/data/v49.0/wave/watchlist/schedules",
 "templates": "/services/data/v49.0/wave/templates",
 "trendedreports": "/services/data/v49.0/wave/trendedreports"
}
```

###


# Einstein.GetDashboards

This method is used to fetch all the available dashboards that you created in Einstein platform.

#### Parameters:

This method does not require any parameter.

#### Sample Response:

```json
{
 "dashboards": [
 {
 "allowPreview": true,
 "assetSharingUrl": "https://ap16.salesforce.com/analytics/wave/dashboard?assetId=0FK2w000000J0fdGAC&orgId=00D2w00000BbXPz&loginHost=ap16.salesforce.com&urlType=sharing",
 "createdBy": {
 "id": "0052w000005ltohAAA",
 "name": "Sandeep Rimal",
 "profilePhotoUrl": "https://c.ap16.content.force.com/profilephoto/005/T"
 },
 "createdDate": "2020-09-04T08:00:54Z",
 "datasets": [
 {
 "id": "0Fb2w000000HGNECA4",
 "label": "Cases",
 "name": "case1",
 "url": "/services/data/v49.0/wave/datasets/0Fb2w000000HGNECA4"
 },
```

###


# Einstein.GetDataSets

This method is used to fetch all the available datasets that are availabled in your Einstein platform acccount.

#### Parameters:

This method does not require any parameter.

#### Sample Response:

```json
{
 "datasets": [
 {
 "clientShardsUrl": "/services/data/v49.0/wave/datasets/0Fb2w000000HGNMCA4/shards",
 "createdBy": {
 "id": "0052w000005ltohAAA",
 "name": "Sandeep Rimal",


 "profilePhotoUrl": "https://c.ap16.content.force.com/profilephoto/005/T"
 },
 "createdDate": "2020-09-04T08:00:50Z",
 "currentVersionId": "0Fc2w000002C0f9CAC",
 "currentVersionUrl": "/services/data/v49.0/wave/datasets/0Fb2w000000HGNMCA4/versions/0Fc2w000002C0f9CAC",
 "dataRefreshDate": "2020-09-04T08:08:56Z",
 "datasetType": "default",
 "folder": {
 "id": "00l2w000000uYhpAAE",
 "label": "VoiceworxSalesApp",
 "name": "VoiceworxSalesApp",
 "url": "/services/data/v49.0/wave/folders/00l2w000000uYhpAAE"
 },
 "id": "0Fb2w000000HGNMCA4",
 "label": "Pipeline Trending",
 "lastAccessedDate": "2020-11-19T12:10:42Z",
 "lastModifiedBy": {
 "id": "0052w000006TsBIAA0",
 "name": "Integration User",
 "profilePhotoUrl": "https://c.ap16.content.force.com/profilephoto/005/T"
 },
 "lastModifiedDate": "2020-09-04T08:08:57Z",
 "lastQueriedDate": "2020-11-19T11:50:00Z",
 "name": "pipeline_trending1",
 "permissions": {
 "create": true,
 "manage": true,
 "modify": true,
 "view": true
 },
 "type": "dataset",
 "url": "/services/data/v49.0/wave/datasets/0Fb2w000000HGNMCA4",
 "versionsUrl": "/services/data/v49.0/wave/datasets/0Fb2w000000HGNMCA4/versions"
 }
```

###


# Einstein.GetLenses

This method is used to fetch all the available lenses that are available in your Einstein platform account.

#### Parameters:

This method does not require any parameter.

#### Sample Response:

```json
{
 "lenses": [
 {
 "allowPreview": true,
 "assetSharingUrl": "https://ap16.salesforce.com/analytics/wave/lens?assetId=0FK2w000000J0fYGAS&orgId=00D2w00000BbXPz&loginHost=ap16.salesforce.com&urlType=sharing",
 "createdBy": {
 "id": "0052w000005ltohAAA",
 "name": "Sandeep Rimal",
 "profilePhotoUrl": "https://c.ap16.content.force.com/profilephoto/005/T"
 },
 "createdDate": "2020-09-04T08:00:51Z",
 "datasets": [
 {
 "id": "0Fb2w000000HGNMCA4",
 "name": "pipeline_trending1",
 "url": "/services/data/v49.0/wave/datasets/0Fb2w000000HGNMCA4"
 }
 ],
 "dateVersion": 1,
 "files": [],
 "filesUrl": "/services/data/v49.0/wave/lenses/0FK2w000000J0fYGAS/files",
 "folder": {
 "id": "00l2w000000uYhpAAE",

 "label": "VoiceworxSalesApp",
 "name": "VoiceworxSalesApp",
 "url": "/services/data/v49.0/wave/folders/00l2w000000uYhpAAE"
 },
 "historiesUrl": "/services/data/v49.0/wave/lenses/0FK2w000000J0fYGAS/histories",
 "id": "0FK2w000000J0fYGAS",
 "label": "Historical Pipeline Stage by Quarter",
 "lastAccessedDate": "2020-11-19T12:10:41Z",
 "lastModifiedBy": {
 "id": "0052w000005ltohAAA",
 "name": "Sandeep Rimal",
 "profilePhotoUrl": "https://c.ap16.content.force.com/profilephoto/005/T"
 },
 "lastModifiedDate": "2020-09-04T08:00:51Z",
 "name": "Historical_Pipeline_Stage_By_Quarter1",
 "permissions": {
 "create": true,
 "manage": true,
 "modify": true,
 "view": true
 },
 "refreshDate": "2020-09-04T08:07:48Z",
 "type": "lens",
 "url": "/services/data/v49.0/wave/lenses/0FK2w000000J0fYGAS",
 "visualizationType": "stackvbar"}
```


# Data Source : Twilio

For  Twilio, ixHello supports

1. SendTextSMS
2. SendSMSWithMedia

#### Send Text SMS

| Parameter | Sample Value | Info                                                                                                                                        |
| --------- | ------------ | ------------------------------------------------------------------------------------------------------------------------------------------- |
|           |              |                                                                                                                                             |
| ToNumber  | {{ToNumber}} | A phone number which is going to receive the message.                                                                                       |
| Body      | {{Message}}  | <p>Actual Message Content.</p><p>Note: {{}} represents that the value for these parameters are going to be read from the user as input.</p> |

Fill the following details:

| Parameter             | Info                                                                                                                                                                                                                                                                                                                     |
| --------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
|                       |                                                                                                                                                                                                                                                                                                                          |
| Method Name           | SendTextSMS                                                                                                                                                                                                                                                                                                              |
| Request Type          | <p><strong>SendTextSMS</strong></p><p>This is the method that you are using</p>                                                                                                                                                                                                                                          |
| Parameters            | As explained in table above                                                                                                                                                                                                                                                                                              |
| Result Type           | <p>Record</p><p>It indicates that the method can return a single record.</p>                                                                                                                                                                                                                                             |
| Record JsonPath       | <p>Each method returns data in json format and to pick any specific information from the json, we need to specify the <em>Record JsonPath</em> so ixHello will pick that data in consideration while displaying results.</p><p>Example: $.\[\*]</p><p>This selects all the records that is returned.</p>                 |
| Empty Result Template | <p>I'm sorry, there was problem delivering your message to {{Input.ToNumber}}. Please try again, How else can I help you?</p><p>Note:</p><p>Anything written in between {{ }} is represented as a json field and it’s case-sensitive. Input.\* refers that you are taking that value from the inputs provided above.</p> |
| Result Template       | Your message has been delivered to {{Input.ToNumber}}. How else can I help you?                                                                                                                                                                                                                                          |

![](/files/gweh1hiCetvCk9CtEVe6)

You can now test the method. On the actions menu, click on Test, provide the parameter value and

Click on Method Test . You can observe the return from method in Object tab and the result displayed to user in Result Tab.&#x20;

<figure><img src="/files/DCNyDZo9gtDUozth9Eet" alt="" width="563"><figcaption></figcaption></figure>

#### Send SMS With Media

| <ol><li>Parameter</li></ol> | Sample Value                                                                           | Info                                                  |
| --------------------------- | -------------------------------------------------------------------------------------- | ----------------------------------------------------- |
|                             |                                                                                        |                                                       |
| ToNumber                    | <p>{{ToNumber}}</p><p>e.g. 6787014967</p>                                              | A phone number which is going to receive the message. |
| Body                        | <p>{{Message}}</p><p>e.g. Hello there</p>                                              | Actual Message Content                                |
| MediaURL                    | <p>{{MediaUrl}}</p><p>e.g\['<https://voiceworx.ai/assets/images/vw-logo-100.png>']</p> | Public URL for the media content.                     |

Fill the following details:

| Parameter             | Info                                                                                                                                                                                                                                                                                                     |
| --------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|                       |                                                                                                                                                                                                                                                                                                          |
| Method Name           | SendSMSWithMedia                                                                                                                                                                                                                                                                                         |
| Request Type          | <p>SendSMSWithMedia</p><p>This is the method that you are using</p>                                                                                                                                                                                                                                      |
| Parameters            | As explained in table above                                                                                                                                                                                                                                                                              |
| Result Type           | <p>Record</p><p>It indicates that the method can return a single record.</p>                                                                                                                                                                                                                             |
| Record JsonPath       | <p>Each method returns data in json format and to pick any specific information from the json, we need to specify the <em>Record JsonPath</em> so ixHello will pick that data in consideration while displaying results.</p><p>Example: $.\[\*]</p><p>This selects all the records that is returned.</p> |
| Empty Result Template | <p>I'm sorry, there was problem delivering your message to {{Input.ToNumber}}. Please try again, How else can I help you?</p><p>Note:</p><p>Anything written in between {{ }} is represented as a json field and it’s case-sensitive.</p>                                                                |
| Result Template       | Your message has been delivered to {{Input.ToNumber}}. How else can I help you?                                                                                                                                                                                                                          |

You can repeat the same method to test the method. Following screen shows a sample response.

![](/files/7HRKWY57yzXRJDjF9T7r)


# AI-Contents

AI Contents act as a data source for the custom app. You can define the behavior, personality, and knowledge base for your Gen AI assistant using AI Contents.

## Types of AI Content

### a. Text Content

You can directly provide text content to define your application, provide information, set rules, expectations and knowledge base.

### b. Training URLs

You can also provide live URLs that will be relevant to your application and which you wish your application to refer to generate responses to your client queries.


# Add AI Content

## Basic Mode

### Gen AI App

When creating a new assistant in Basic Mode, you can provide text content and Web URLs within the assistant creation wizard.

<figure><img src="/files/wgqZC0EaCUzq5G2eW3UM" alt=""><figcaption></figcaption></figure>

You can provide text content in ‘Additional Information – Text Content’ section.

You can add additional training URLS in ‘Additional Information – Website URLs’ section. Every URL you add will be equivalent to a new AI-Content.

The new assistant created uses these as AI-Content inputs for training.

### Non-Gen AI-App

For Non-Gen AI-Apps, when creating a new assistant, there is no explicit ‘Training’ section, but a sample text content is created, which can later be modified in Studio Mode.

## Studio Mode

From the left navigation, click on Apps-> Custom Apps to view the list of assistants.

Select any assistant for which you want to create AI-Contents.

<figure><img src="/files/JkGNnjieEADuVtOUsuDi" alt=""><figcaption></figcaption></figure>

Go to AI Contents tab. The list displays already created contents under the pattern ‘Content\_Categoy / AIContent\_Name’. In the example below, ‘ACME HVAC AssistantXXX’ is the category and ‘ACME HVAC Assistant\_Train’ is the content name.

<figure><img src="/files/O3RkO6RfXBUrQAx2ziWB" alt=""><figcaption></figcaption></figure>

To add new AI content, Click on Add New AI Content button.

<figure><img src="/files/FI4cxlh1hUgQQNbr69tM" alt=""><figcaption></figcaption></figure>

**Add Training URLs**

You can directly add Training URLS, and for text content, you can select from available list of text content from the drop-down menu and click on Save.

**Add New Text Content**

If there are no available text contents, you can add one by clicking ‘Add new’.

<figure><img src="/files/K1IbSLHoc6ggj3apEwgg" alt=""><figcaption></figcaption></figure>

On clicking ‘Add New’ in the above interface, you are redirected to Content-> Content menu.

<figure><img src="/files/gA5DQ1BVBM01tlfCGW2K" alt=""><figcaption></figcaption></figure>

Click on Add New Content. You can refer to this document on how to create a Content.

<figure><img src="/files/KuWm0AYYeMSEwyuEELki" alt=""><figcaption></figcaption></figure>

While creating the content, make sure that you select the same category as your app category. You can find your app category from the custom app list.

<figure><img src="/files/tkKRY8DXJK2sphw3C703" alt=""><figcaption></figcaption></figure>

Once you’ve created the content, now let’s go back to the AI Contents tab from Custom Apps and click on Add new AI Content.

<figure><img src="/files/5jODEng1pNbYcKvSe0Ww" alt=""><figcaption></figcaption></figure>

The newly created AI Content shall be available in the drop down for Training Content. Select that and click on Save.

<figure><img src="/files/ASn91EBOAEpDpWRIdmi1" alt=""><figcaption></figcaption></figure>

The newly added AI-Content is now available on the list.

<figure><img src="/files/JxAWgQjRKX4urhOEVeZM" alt=""><figcaption></figcaption></figure>

**Alternative Method**

Go to Content -> AI-Content.

Select the correct category in the category drop down and click on Add New Content.

<figure><img src="/files/ci3ZAtMflWA2ExKYw9iL" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/RniaoayEnLWEFAKnE9Kc" alt=""><figcaption></figcaption></figure>

Find the added content in ‘Select Training Content’ drop down under ‘Text Content’ tab. Add Web URLs in the Web URLs tab and click on Save.

The AI Content is now mapped to the app and will be used for training the app.

<figure><img src="/files/9Mv4f8n3dmK6rYG1InzL" alt=""><figcaption></figcaption></figure>


# K-BOT Integration

### Overview:&#x20;

Users are now able to utilize knowledge bots from
&#x20;iX Hello Customer and Employee Facing screens.

### Prerequisites:

* **An iX Hello Account** For steps to [create an iX Hello account,](https://app.gitbook.com/o/-M8Qw0HjmL3rDRYUXBX0/s/-M8XHvUsfyTUFLvToHqD/general/ix-hello-create-account) follow the steps at
* Use this link to [Sign up for iX Hello platform](https://bots.ixhello.com/Account/Login?ReturnUrl=%2F).
* To create an APP using Quick Start App refer to [this](https://app.gitbook.com/o/-M8Qw0HjmL3rDRYUXBX0/s/-M8XHvUsfyTUFLvToHqD/general/rag-file-upload/file-upload-in-studio-mode-1/quick-start-apps)
* To create an App using Blueprint refer to [this](/ixhc/blueprints/create-app-using-app-blueprints)
* To Access the K-BOT platform, users account should be setup on the respective K-Bot platform.

### **How to connect to K-BOT using iX Hello:**

To connect to K-BOT there are two different approaches,

#### Step 1: **Using the Blueprint**

Let's see how to create a K-BOT using the Blueprint.

**"App Blueprints"** section of the **ixHello platform**, where users can browse and select pre-configured templates to quickly build AI-powered applications.

Each blueprint includes a title, description, and options to view more details or create the app.

Now click on Create App on Knowledge Bot. This allows users to **quickly launch AI apps** using ready-made templates tailored for specific use cases.

<figure><img src="/files/FHrJJHwjhBC3re2V6ZM3" alt="" width="563"><figcaption></figcaption></figure>

Follow the rest of the steps as mentioned below.

#### Step 2: **Using the Quick Start Apps**

Let's see how we can create a K-BOT using QuickStart Apps:

In Detail to know about QuickStart App refer to this [document](/ixhc/general/rag-file-upload/file-upload-in-studio-mode-1/quick-start-apps).

On the Home Screen, click on the Quick Start App Button.

<figure><img src="/files/2b154KSwmLdaKE86DMxt" alt="" width="563"><figcaption></figcaption></figure>

You will be navigated to Quick Start Apps screen and then click on "Add a New Assistant" button.

<figure><img src="/files/ZHJpM5dUJMuGPrUnfi8f" alt="" width="563"><figcaption></figcaption></figure>

You will see the Quick Start Apps Library, which helps you to create an app using ready to use AI apps.

<figure><img src="/files/sVEpZ73pvOlSFqjlY3wl" alt="" width="563"><figcaption></figcaption></figure>

Now using this ready to use app, you can easily create a K-Bot.&#x20;

1. &#x20;**Let's create a K-BOT using the ready to use AI apps by clicking on the K--BOT Create App button.**

Once clicked on the create app, you will need to fill in all the mandatory sections under, General Information and Knowledge Bot tabs.

* **General Information (Tab)**
* **Name:** Field where you specify the **name of your assistant**.
* **Description:** Field to describe the purpose of your assistant.
* **Welcome Message:** The **greeting message** shown to users when they first start interacting with the assistant.
* **Goodbye Message:** The **closing message** shown when the interaction ends.

<figure><img src="/files/B0mdiSNMFMtCzWUl8fwD" alt="" width="563"><figcaption></figcaption></figure>

2. When you click on the "Knowledge Bot" tab, you'll see a dropdown list of available bots.&#x20;

* **Knowledge Bot (Tab)**
* **Bot Names:** The available **Knowledge Bots** you can connect. You select one of these example apps to associate it with your app.
* **Click Save Button**

<figure><img src="/files/zQzqzl0MPUjC7hVMPFWS" alt="" width="563"><figcaption></figcaption></figure>

3. If you do not want to use the existing K-Bot, **if you wish to create a new K-Bot, then click on Create Button.**

<figure><img src="/files/7Ayv8bTGgpspNNKJYfFD" alt="" width="563"><figcaption></figcaption></figure>

4. When you click on create button, you will be navigated to K-Bot Login Screen, click "Login with Microsoft".

<figure><img src="/files/BK3IMj0IGL1aCXIW5yB7" alt="" width="563"><figcaption></figcaption></figure>

5. You need to provide your SSO ID to Login to K-Bot Creation Page.

<figure><img src="/files/kE6NwIM6gpsyGZPya3Ve" alt="" width="563"><figcaption></figcaption></figure>

6. **On the Landing Screen, To Edit the Existing K-BOT**&#x20;
   1. **Click on the Pencil Icon** → Edit the bot’s details (name, description, settings, etc.).&#x20;
   2. Update the necessary Information and save, the screenshot below is self-explanatory

<figure><img src="/files/xSehSmoEF9cKHAh7Kq6x" alt="" width="393"><figcaption></figcaption></figure>

#### **Step 3: Create a New KBOT**

1. To Create a new K-BOT then click on "Add Knowledge Bot" follow the same steps as above, fill in all the necessary fields and click on save button.

<figure><img src="/files/BF2TGHc174e2iF9tZx1a" alt="" width="264"><figcaption></figcaption></figure>

2. The Training data which is added to the newly created bot are in the form of Document and the URL.

<figure><img src="/files/c6xbDUbEChoskiSiOX9r" alt=""><figcaption></figcaption></figure>

3. Once the DataSource's are added then the newly created bot will appear in the app list as shown below.

<figure><img src="/files/G410iQYdGu9oXd3v3hdE" alt=""><figcaption></figcaption></figure>

4. **Now let's get into the iX Hello and see how we can configure the KBOT.**
   1. **In configuration screen** within the **"Knowledge Bot" tab** of a K-BOT application setup interface. You are in the process of **connecting a knowledge Bot** to your application using the platform **ixHello**. Choose the K-BOT which you have created in K-BOT platform.
   2. Click Save.

<figure><img src="/files/CFvW94ciLUMELTo8tiGV" alt="" width="563"><figcaption></figcaption></figure>

5. Once saved, the app will appear in the section of the **Quick Start Apps** interface, specifically under **"My Assistants"**, where users can manage and activate different types of interactions for a selected assistant.
6. Click on **"Activate"** button, allowing you to enable that specific channel:
   1. **Webchat**: For browser-based conversational interaction.
   2. **Digital Human**: Likely a visual avatar or human-like interface.
   3. **Call/Text**: For voice or SMS-based communication.

<figure><img src="/files/IJUje8sxEqQsJm8WZ9hl" alt="" width="563"><figcaption></figcaption></figure>

7. This interface is designed to let you quickly deploy and manage AI assistants across multiple channels.
8. Now let's go-ahead and Converse with then K-BOT which we created.

{% hint style="info" %}
The Bot from iX Hello will interact with the K-BOT which is created on the K-BOT interface and get the response based on the Knowledge base available on the K-BOT interface.
{% endhint %}

<figure><img src="/files/yUBlKXTxY7oqLJIYaLd2" alt="" width="375"><figcaption></figcaption></figure>

9. The bot has used the knowledge from the K-Base to answer the query:
   1. Sample Training Data:

<figure><img src="/files/ehCvh0bJSjOg2c85dQvo" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/dSPF3l8F9NbwglKPlQS3" alt="" width="375"><figcaption></figcaption></figure>

10. In the above the response the bot has used the URL provided -
    1. [https://www.concentrix.com/](<&#xD;&#xA;https://www.concentrix.com/>) and [
       ](<&#xD;&#xA;https://www.concentrix.com/about/>)
    2. [https://www.concentrix.com/about/](<&#xD;&#xA;https://www.concentrix.com/about/>)

### **Step 4: Access the bot in** iX Hello Employee

To enable to bot to access in iX Hello Employee, need to follow the below mentioned simple steps:

1. The Bot, which was created earlier using quick start app, the same app will also be visible in the custom apps.

<figure><img src="/files/k0GBbtW0Y0Fuss7MN93T" alt=""><figcaption></figcaption></figure>

2. Select “**Publish**” in the top menu tabs then click “**Add App Publishing**”. Then add the below details:

| External Skill Name | Demo K-Bot                                                                                                                                                                                                                                                                |
| ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| App Description     | You are a concise and focused assistant, providing accurate information based solely on the provided Content files from Document attached and the URLs provided. Aim to deliver clear and relevant answers to your queries from Content files from Document and the URLs. |
| Skill Type          | Search                                                                                                                                                                                                                                                                    |
| Channel             | Gen AI                                                                                                                                                                                                                                                                    |
| Toggle Button       | Enable the API-Access toggle button                                                                                                                                                                                                                                       |

3. Click Save.
4. To update settings for an existing flow, click the **more icon** **button** and select **Settings** from the dropdown.
5. If you want to see this custom app in the Basic mode: **Enable** the **Show this app in the Basic Mode** toggle button.
6. **Optional**: Replace all placeholder values in all “Message”. Uncheck all the settings shown in the second image below.

<figure><img src="/files/hZdN5N7Fp6siwV5Xcs9S" alt=""><figcaption></figcaption></figure>

Once your K-Bot is created and configured, it’s ready to deliver intelligent, context-aware responses across multiple channels. With flexible PII redaction, customizable knowledge sources, and seamless integration options, your K-Bot is now equipped to enhance user engagement and streamline information delivery.

After enabling API access, users can interact with the bot via the iX Hello Employee platform. Refer to this [document](https://app.gitbook.com/o/-M8Qw0HjmL3rDRYUXBX0/s/l2dmTEKSqnEOjqqng3nd/~/changes/176/integration/k-bot-integration) for understanding bot interactions.


# Identify RAG Bots as Knowledge Bots

Here we are essentially classifying bots that use **Retrieval-Augmented Generation (RAG)** as part of the **Knowledge Bot category** within a platform (like iXHello). This is because their core purpose aligns with what Knowledge Bots are meant to do: **provide accurate, context-aware answers based on enterprise knowledge sources**.

#### **Why RAG Bots = Knowledge Bots**

* **Knowledge Bots** are designed to deliver **trusted, data-backed responses**.
* **RAG bots** achieve this by:
  * **Retrieving** relevant information from enterprise knowledge bases, documents, or APIs.
  * **Generating** a natural language response using an LLM, grounded in retrieved content.
* This makes them fundamentally different from generic chatbots (which rely only on the model’s memory) and task bots (which execute workflows).

Lets see how we can use RAG Bots as K- Bots

### Quick Start App Guide - Studio Mode

#### Prerequisites

How to use [Quick Start App](/ixhc/general/rag-file-upload/file-upload-in-studio-mode-1/quick-start-apps) use this document

How to create a KBOT use [this](/ixhc/general/k-bot-integration)

#### RAG Blueprint Selection

When a user selects a RAG Blueprint in the Quick Start app, an option checkbox will appear with the ability to mark it as a Knowledge Bot.

#### Knowledge Bot Transformation

By selecting the Knowledge Bot option, the RAG-Bot can be classified as a Knowledge Bot. This change enables the bot to be utilized by employees or other teams using our APIs.

<figure><img src="/files/7uEiKlLw9CoQoF9J0oCJ" alt="" width="563"><figcaption></figcaption></figure>

Access the KB contents in Custom App mode by navigating to AI contents.

<figure><img src="/files/DZEZPfPOvBMQWFlsPFgb" alt=""><figcaption></figcaption></figure>

To make a bot a KBOT, check the KBOT checkbox when creating an app. Alternatively, edit existing apps in the Custom App list.

* In **Custom Apps**, users can upload files under **AI Content** to provide a knowledge base for the bot.
* This means the bot can be configured to **retrieve and use enterprise knowledge** for answering queries.
* Therefore, the option to **mark the bot as a Knowledge Bot** is shown here, even if the user hasn’t uploaded files yet.
* This gives flexibility:

  * Users can decide upfront that the bot will act as a Knowledge Bot.
  * Later, they can upload knowledge files to enable that functionality.

  <figure><img src="/files/ldAHFv5SGrJgoat1IdLn" alt="" width="563"><figcaption></figcaption></figure>

### Knowledge Bot Configuration in Basic Mode

* In **Basic Mode apps**, the **Knowledge Bot checkbox** is visible on the **Data tab** of an existing app. Alternatively, you can create a new app using any blueprint, where the Kbot checkbox will also be available.
* This checkbox allows you to **mark the app as a knowledge Bot**, even if you haven’t uploaded any training files yet.
* The reason: **Basic Mode supports adding training files later**, so the option is kept for flexibility.

<figure><img src="/files/mlw8KioxEzE0x505VaF4" alt="" width="563"><figcaption></figcaption></figure>


# PII-Management

### **Overview**

**PII Redaction in Ix Hello Customer** refers to the process of automatically identifying and concealing **Personally Identifiable Information (PII)** from customer interactions—especially in call recordings, transcripts, and other data formats—to ensure privacy, security, and regulatory compliance.

PII redaction involves **removing or masking sensitive personal data** such as:

* Names
* Phone numbers
* Email addresses
* Credit card details
* Social Security numbers
* Birthdates
* Addresses

This ensures that even if data is accessed, **personal details remain protected**.

### &#xD;Benefits:

• Increased overall Data Security and control.
\
• Meets sensitive data compliance concerns
\
• Manageable LLM data use & interactions
\
• Increased Org control of bot builders data exposure

#### **Why Is It Important?**

* **Compliance**: Helps meet regulations like **GDPR**, **CCPA**, and **PCI DSS**.
* **Customer Trust**: Shows commitment to data privacy, enhancing customer confidence.
* **Risk Reduction**: Minimizes chances of **data breaches** and **identity theft**.
* **Operational Efficiency**: Automates a time-consuming process, reducing manual errors.

### Let's learn how to configure PII Redaction in iX Hello:

*We can enable PII Redaction in both Basic Mode and as well as Studio mode.*

### Step 1: Studio Mode:

Login to the iX Hello application. Under Studio mode choose the organization menu. With in Organization menu choose ("Manage Organization Configurations") **Administration** menu.

These settings determine how **Personally Identifiable Information (PII)** is handled in bots created within the organization.

**1. Mandate PII Configuration**

* If enabled:
  * Bot creators **must configure PII redaction**.
  * Alternatively, they must **explicitly acknowledge** that no PII is present in the bot.
  * This enforces stricter compliance and accountability.

**2. PII Configuration is optional**

* If enabled:

  * Bot creators **are not required** to configure PII redaction.
  * However, they **can still choose** to configure it if needed.
  * Offers flexibility but may pose a risk if PII is unintentionally left unredacted.    &#x20;

  &#x20;         &#x20;

  <figure><img src="/files/RBvEqt21SIDqs6zSrGvx" alt=""><figcaption></figcaption></figure>

### Step 2: Basic Mode:

#### How to enable PII Redaction in Basic Mode Apps:

1. Login to iX Hello in Basic Mode, specifically under the **"My Assistants"** section.
   1. Dropdown Menu Options: **PII Redaction:** A feature to remove personally identifiable information from conversations or data.
   2. Click on the PII Redaction.&#x20;

<figure><img src="/files/XsO9KNcEcyQe1VV3Bn9J" alt="" width="563"><figcaption></figcaption></figure>

2. Once clicked on PII Redaction, PII Redaction settings pop-up for the chosen assistant within the "My Assistants" section of a platform interface.
   1. Toggle Option: You can enable or disable automatic redaction of PII. In the image, it's currently turned off.&#x20;
   2. Disclaimer:
      1. Impact: It warns that enabling redaction might reduce the bot’s accuracy and performance, since important context could be hidden.

<figure><img src="/files/PxYBoRFMDKHOUSr747SF" alt=""><figcaption></figcaption></figure>

3. If you want to enable the PII Redaction, click on the toggle button and enable it, once enabled, you encounter the below screen and You can choose to redact any of the following data types:

<table><thead><tr><th width="220">Setting</th><th>Description</th></tr></thead><tbody><tr><td>Credit Card</td><td>12–19-digit card numbers</td></tr><tr><td>Crypto wallet</td><td>Cryptocurrency wallet addresses (e.g., Bitcoin).</td></tr><tr><td>Date/Time</td><td>Specific or relative time references.</td></tr><tr><td>Domain Name</td><td>Internet domains like <code>example.com</code>.</td></tr><tr><td>Email Address</td><td>Identifiable email addresses.</td></tr><tr><td>IBAN Code</td><td>International bank account numbers.</td></tr><tr><td>IP Address</td><td>IPv4 or IPv6 addresses.</td></tr><tr><td>Location</td><td>City, region, or geographic names.</td></tr><tr><td>National Registration Plate</td><td>Vehicle license plate numbers.</td></tr><tr><td>Person Name</td><td>Names identified through context and models.</td></tr><tr><td>Phone Number</td><td>With or without country codes.</td></tr></tbody></table>

4. You can choose all the options or choose few based on the client's requirement which you want to mask or hide it from the users.

<figure><img src="/files/KMuMf8wnrvlpedVKBRGk" alt=""><figcaption></figcaption></figure>

5. Once clicked save then all the information will be hidden from the users, and the bot will be PII enabled.

<figure><img src="/files/TcILWEB073O8U6qzUW8g" alt=""><figcaption></figcaption></figure>

6. The same app where PII redaction was enabled in Basic Mode can also be accessed in Studio Mode, allowing you to enable or disable PII redaction based on your specific use case — and vice versa.
7. Open the Custom App and navigate to the "Publish" tab. If PII redaction is enabled, you'll see a green shield icon. If it's disabled, the shield icon will appear grey.

<figure><img src="/files/PappThJj5s9Pi3meXQFG" alt=""><figcaption></figcaption></figure>

8. If you want to disable the PII redaction in the existing app in the studio mode, then navigate to Publish tab and click on the App Name and you will see the toggle button to disable it.

<figure><img src="/files/Qf2B6HVqxFvB0HHfSxt3" alt=""><figcaption></figcaption></figure>

9. Once you click on save button, the changes will be saved and the PII redaction will be disabled to that particular app, and the Shield Icon will turn to grey.

<figure><img src="/files/DPQSjUvblhQDO6trpwh8" alt=""><figcaption></figcaption></figure>

### Step 3: You can even configure PII Redaction from different Tabs in Studio Mode:

1. AI-Content: Navigate to AI content tab and then click on Add New Content button, you will encounter Adding AI Content through file upload Settings:

<figure><img src="/files/QMCbdJIQvMJuidTRQKSM" alt=""><figcaption></figcaption></figure>

2. In this particular app, when uploading content, **PII Redaction** will automatically be applied to the following entities:
   1. **CREDIT\_CARD**: Credit card numbers
   2. **CRYPTO**: Cryptocurrency wallet addresses
   3. **DATE\_TIME**: Dates and times
   4. **DOMAIN\_NAME**: Internet domain names
   5. **EMAIL\_ADDRESS**: Email addresses

This ensures sensitive information is masked during AI content processing.

3. Method Screen: Navigate to Method tab and then click on Edit action button of the existing method, you will encounter a configuration screen for the input method, where in you will see the PII Fields,
   1. **PII\_Entities**:\
      You can specify which types of PII to redact (e.g., `CREDIT_CARD, EMAIL_ADDRESS`).
   2. **PII\_Language**:\
      Defines the language context for PII detection (e.g., `en`, `fr`).

<figure><img src="/files/tPLNUh9Q1iKCQMakOOPE" alt=""><figcaption></figcaption></figure>

4. Leaving these fields blank means **PII redaction is not currently configured**, which could pose a privacy risk depending on the use case.
5. Let's see how the bot responds using PII redaction and without PII Redaction.
6. With Out Enabling PII Redaction the response from the Bot is:

<figure><img src="/files/pxLtGYfpPgV1HGg3xOT4" alt="" width="375"><figcaption></figcaption></figure>

7. With Enabling the PII Redaction specifically for " Location" the Bot response will be as below:

<figure><img src="/files/pVgfcHG06i8RgkYkV6gv" alt="" width="375"><figcaption></figcaption></figure>

Now you know how to use PII redaction in the apps.


# Retrospective Redaction User Guide

### Overview

The Retrospective Redaction feature enables administrators to identify and anonymize sensitive information within historical conversation data that has already been stored in the system.

While Anonymization Strategies protect data during document ingestion and future processing, Retrospective Redaction focuses on previously stored conversation records, ensuring that historical data also complies with privacy, security, and regulatory requirements.

With this enhancement, users can schedule, manage, monitor, and execute redaction jobs directly from the application interface without requiring access to AWS services or manual backend configuration.

***

### Key Benefits

* Protect historical conversations containing sensitive information.
* Schedule automated redaction jobs.
* Execute redaction immediately using the Run Now option.
* Monitor schedule status and execution results from a centralized dashboard.
* Reduce dependency on backend or cloud service configuration.
* Help organizations meet privacy and compliance requirements.

***

### Accessing Retrospective Redaction

A dedicated **Retrospective Redaction** option is available in the application navigation menu.

The Retrospective Redaction page provides:

* Existing schedules
* Create Schedule functionality
* Search and filtering options
* Schedule status tracking
* Execution history
* Processing information
* Schedule management actions

<figure><img src="/files/QTJQVs2oYBU548t9WJYQ" alt=""><figcaption></figcaption></figure>

***

### Role-Based Access

Access to create and manage redaction schedules is controlled by user roles.

| Role           | Access Level                                                         |
| -------------- | -------------------------------------------------------------------- |
| Org Admin      | Can schedule redaction for all Assistant IDs across the organization |
| Platform Admin | Can schedule redaction for all Assistant IDs across the organization |
| Support Admin  | Can schedule redaction for all Assistant IDs across the organization |
| Employee       | Can schedule redaction only for Assistant IDs they own               |

***

## Creating a Redaction Schedule

### Steps

1. Navigate to Retrospective Redaction.
2. Click Create Schedule.
3. Select an Assistant ID.
4. Choose a Schedule Type.
5. Configure the required schedule details.
6. Save the schedule.

Once saved, the schedule becomes available in the dashboard and can be monitored through the schedule list.

<figure><img src="/files/5gWLp7k1Zvf5kZFQyeSA" alt="" width="563"><figcaption></figcaption></figure>

***

### Schedule Configuration

#### Assistant ID

The Assistant ID identifies the assistant whose conversation history will be processed.

Only authorized Assistant IDs are available based on the user's role and permissions.

***

#### Target Function

The Target Lambda Function is displayed as a **read-only field** and does not require user modification.

This function is automatically used to execute the redaction process.

***

### Schedule Types

#### One-Time Schedule

A One-Time Schedule executes the redaction process once at a specific date and time.

**Configuration Requirements**

* Select the execution date.
* Select the execution time.
* The selected date and time must be in the future.
* The system automatically converts the selected time to UTC before scheduling execution.

<figure><img src="/files/EtJVfcjIUPtrkK0eS1sJ" alt="" width="563"><figcaption></figcaption></figure>

<figure><img src="/files/tnRX73HRdWQ393p9epFy" alt="" width="563"><figcaption></figcaption></figure>

Once the schedule is created, it will be listed in the table of schedule redaction of conversation history.

<figure><img src="/files/Zr0RR8OJviF24pxXIOXC" alt=""><figcaption></figcaption></figure>

When you click on the details , the Details link allows administrators to view the outcome of a redaction job after it has been executed. It provides execution-specific information about the scheduled redaction process.

<figure><img src="/files/Dyv4TxbHnwBy2F4axY4F" alt=""><figcaption></figcaption></figure>

**Recommended Use Cases**

* Compliance audits
* Historical data cleanup
* Immediate privacy remediation activities
* One-time data governance initiatives

***

#### Recurring Schedule

A Recurring Schedule executes the redaction process repeatedly based on a configured frequency.

**Examples**

* Every 15 minutes
* Every hour
* Daily
* Weekly
* Monthly

Recurring schedules are useful when conversation data is continuously generated and regular redaction is required.

**Recommended Use Cases**

* Continuous compliance monitoring
* Ongoing data privacy enforcement
* Automated maintenance of conversation history

***

#### Custom Cron Schedule

For advanced scheduling requirements, users can configure a custom AWS Cron expression.

This option provides greater flexibility for organizations that require complex scheduling patterns beyond standard recurring intervals.

***

### Run Now

The **Run Now** option allows users to execute redaction immediately without waiting for the configured schedule.

<figure><img src="/files/5LoEEVbfR8ufB4jks5Jz" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/4XgPORwnTDIsAkGytvHC" alt=""><figcaption></figcaption></figure>

Once run now action is completed, the status changes to completed.

<figure><img src="/files/czyzpm3dfnqJ5nMuO3Ql" alt=""><figcaption></figcaption></figure>

When you click on the details, the Details link allows administrators to view the outcome of a redaction job after it has been executed. It provides execution-specific information about the scheduled redaction process.

<figure><img src="/files/W681znDkBsyXCDbpe6Pg" alt="" width="563"><figcaption></figcaption></figure>

#### Use Cases

* Urgent redaction requests
* Validation and testing of redaction configuration
* Immediate cleanup of newly identified sensitive information

***

## Redaction Processing

When a scheduled or manual redaction job is triggered, the system performs the following actions:

1. Identifies the selected Assistant.
2. Retrieves conversation history associated with the Assistant.
3. Filters records based on the configured processing scope.
4. Detects sensitive information.
5. Applies the configured anonymization or redaction strategy.
6. Updates the redacted conversation records.
7. Generates an execution log.
8. Stores the processing results for monitoring and reporting.

***

## Monitoring Redaction Jobs

The Retrospective Redaction dashboard provides visibility into scheduled and completed executions.

Administrators can track:

| Field             | Description                              |
| ----------------- | ---------------------------------------- |
| Schedule Name     | Name of the configured schedule          |
| Assistant ID      | Assistant being processed                |
| Status            | Current schedule or execution status     |
| Scheduled Time    | Configured execution time                |
| Next Run Time     | Next scheduled execution                 |
| Process Count     | Number of conversations processed        |
| Error Details     | Any processing errors encountered        |
| Execution History | Historical execution details and results |

***

### Search and Filtering

The dashboard includes search and filtering capabilities, allowing users to:

* Locate schedules quickly.
* View schedules by status.
* Filter schedules by Assistant ID.
* Review completed or pending executions.

***

## Execution Results

After a redaction job completes, execution details are available from the dashboard.

Users can review:

* Processing status
* Number of records processed
* Success or failure details
* Execution timestamp
* Associated Assistant ID

This information helps administrators verify that historical data has been successfully redacted.

***

#### Schedule Time Validation

For one-time schedules:

* The selected date and time must be in the future.
* Invalid dates or past timestamps cannot be scheduled.

***

#### Assistant Ownership Validation

Employees can create schedules only for assistants they own, while administrative roles can manage schedules for all assistants within the organization.

***

## Conclusion

The Retrospective Redaction feature provides a centralized and user-friendly way to anonymize sensitive information within historical conversation data. By enabling scheduled and on-demand redaction directly from the application, organizations can maintain data privacy, support compliance requirements, and ensure that both historical and future conversation records remain protected.


# Anonymization Strategies

### Overview

The Anonymization Strategy feature helps protect sensitive information by transforming Personally Identifiable Information (PII) before it is stored and used by AI-powered applications.

When documents are uploaded to the AI Content screen, the selected anonymization strategy is applied automatically. The protected content is then processed, indexed, and stored for retrieval.

This ensures that AI responses are generated using anonymized data rather than exposing original sensitive information.

***

### Supported Anonymization Strategies

The platform supports the following anonymization methods:

| Strategy                           | Description                                                           |
| ---------------------------------- | --------------------------------------------------------------------- |
| Replacement (Default)              | Replaces sensitive information with redacted placeholders.            |
| Masking                            | Hides sensitive data using masking characters such as asterisks (\*). |
| Encryption                         | Encrypts sensitive data before storage.                               |
| Format-Preserving Encryption (FPE) | Encrypts data while maintaining the original format and structure.    |
| Hashing                            | Converts sensitive information into irreversible hash values.         |
| Custom Scrambler                   | Rearranges characters or digits to obscure the original value.        |
| Pseudonymization                   | Replaces sensitive values with realistic substitute values.           |

{% hint style="info" %}
Note: If no strategy is selected, the system automatically applies the Replacement strategy.
{% endhint %}

***

### Where to Configure Anonymization Strategies

Anonymization strategies can be configured within:

* Basic Mode
* AI Content
* Publish Configuration

The selected strategy is applied during document processing.

***

### Applying an Anonymization Strategy

#### Steps

1. Navigate to the AI-Content screen.

<figure><img src="/files/KQ75ErZnJg9fSuasEpsB" alt=""><figcaption></figcaption></figure>

Select the desired Anonymization Strategy out of 7 available strategies,

<figure><img src="/files/aRTMMhqEVsQD4lOyfMjU" alt="" width="563"><figcaption></figcaption></figure>

Click Save.

Upload the document.

<figure><img src="/files/mbTQAr0YL9A7petR0bqU" alt="" width="563"><figcaption></figcaption></figure>

<figure><img src="/files/TqvkahHUBFLUNwh6hsxE" alt=""><figcaption></figcaption></figure>

Publish the content and test the bot.

<figure><img src="/files/8laen14aXGcPqPfiHY0Z" alt=""><figcaption></figcaption></figure>

Query the bot using information from the uploaded document.

<figure><img src="/files/qrquEhFnS4afXzbl68S8" alt="" width="563"><figcaption></figcaption></figure>

The bot will return responses based on the anonymized version of the content.

<figure><img src="/files/Dfjt6rPfYpe2TMzXYdFk" alt="" width="375"><figcaption></figcaption></figure>

***

### Example: Masking Strategy

#### Original Content

Plain Text1Name: John Smith2Phone Number: 9876543210Show more lines

#### Anonymized Content

Plain Text1Name: J\*\*\* S\*\*\*\*2Phone Number: \*\*\*\*\*\*3210

<figure><img src="/files/AvnNWXpxOiDpF7tv5btB" alt="" width="375"><figcaption></figcaption></figure>

***

### Benefits

* Protects existing conversation history.
* Supports privacy and regulatory requirements.
* Eliminates manual redaction efforts.
* Provides automated scheduling options.
* Ensures historical data remains protected.

***

### Conclusion

The Anonymization Strategies feature provides a flexible and effective approach to protecting sensitive information throughout the AI knowledge ingestion and retrieval process. By offering multiple anonymization methods such as Replacement, Masking, Encryption, FPE, Hashing, Custom Scrambler, and Pseudonymization, organizations can choose the strategy that best aligns with their security, privacy, and compliance requirements.

When documents are uploaded, the selected strategy is automatically applied before content is processed, indexed, and stored. This ensures that sensitive information remains protected while still enabling users to leverage AI-powered search and conversational experiences.

By safeguarding data at the source, Anonymization Strategies help organizations enhance data privacy, reduce the risk of exposing confidential information, support regulatory compliance, and maintain trust in AI-driven interactions.


# LLM-Moderation and Safety Layer - Phase 1

### Overview

An LLM-Moderation and Safety Layer is a system designed to ensure that large language models (LLMs) function within safe, ethical, and compliant norms. Acting as a buffer between the model and users, it prevents outputs that may be harmful, biased, or violate policies.

**iX Hello**→ **Moderation & Safety Layer** → **LLM Inference Services**

This layer functions as a **gatekeeper**, implementing:

* **Pre-Generation Filters**: Validates prompts before sending them to the LLM.
* **Post-Generation Filters**: Scans responses before delivery.

#### Why LLM?

LLMs fill critical gaps left by traditional language systems and unlock advanced capabilities for modern businesses. Here’s why they matter:

1\. Master Complex Language Tasks\
Human language is rich, nuanced, and context dependent. Unlike rule-based or small NLP models, LLMs can interpret intent, tone, and context, making interactions more natural and accurate.

2\. Scale Across Multiple Domains\
A single LLM can power customer support, content creation, coding assistance, knowledge retrieval, and more eliminating the need for separate models for each task.

3\. Boost Productivity\
LLMs automate repetitive tasks like document summarization, email drafting, and FAQ responses, saving time for professionals in legal, finance, healthcare, and product management.

4\. Enable Smarter Decisions\
They process vast amounts of text quickly, extract actionable insights, and help businesses make data-driven decisions faster.

5\. Adapt and Continuously Improve\
LLMs can be fine-tuned for specific industries, compliance needs, and learn from feedback to enhance accuracy, safety, and relevance.

#### Purpose of LLM Moderation : Safety and Compliance Features for AI Systems

**1. Content Filtering:**\
Utilize classifiers or rule-based systems to detect and block unsafe outputs like hate speech, sexual content, violence, and misinformation, by effectively flagging high-risk responses.

**2. Policy Enforcement:**\
Ensure alignment with organizational or regulatory guidelines by applying custom rules in sensitive domains such as finance, healthcare, and politics.

**3. Compliance and Legal Safety:**\
Prevent sharing of Personally Identifiable Information (PII) or confidential data. Ensure adherence to GDPR, HIPAA, and organizational security standards.

**4. Protect Brand Reputation:**\
Maintain a professional tone and prevent embarrassing or harmful outputs to build trust through safe and respectful communication.

**5. Improve Accuracy and Reliability:**\
Flag hallucinations or misleading information before reaching users to maintain trustworthy responses in business workflows.

**6. Integration:**\
Intercept user inputs and model outputs before final response delivery. Provide API hooks or middleware for easy integration across services.

**7. Contextual Safety:**\
Understand user intent and context to prevent misuse such as phishing or fraud. Use dynamic risk scoring based on conversation flow.

**8. Human-in-the-Loop:**\
Escalate flagged cases to human moderators for review and provide override mechanisms for critical decisions.

**9. Audit and Logging:**\
Maintain logs for compliance and forensic analysis, enabling transparency for regulatory audits.

#### How LLM **Moderation & Safety Layer** acts in iX Hello

#### Moderation Layer Overview

The moderation layer sits between the user interface and the LLM inference service. Its key functions are:

* Reviewing incoming prompts and outgoing model responses.
* Using machine learning and rule-based filters.

The moderation system can:

* Allow, redact, warn, or block content.
* Integrate smoothly with current workflows.
* Support custom policies.
* Log all actions for auditing and analysis.

#### Demo App Overview

Our demo app leverages Azure's AI Content Safety to moderate inputs across four categories: self-harm, violence, sexual content, and hate speech.

#### Chatbot Implementation

Each message sent in the chat is moderated by the language model. Messages that pass the moderation check proceed with the flagged status.

#### Content Moderation

Inputs categorized under self-harm, violence, sexual content, or hate speech are automatically flagged.

#### Alert System

If inappropriate content is detected, a pop-up alert will notify users with the message "Content flagged by moderation."

This preliminary version of the app focuses on basic functionality, marking random inputs not in the flagged categories as safe.

{% hint style="info" %}
Flag Handling in the Application: Currently, flag management is controlled through app secrets, accessible only to authorized personnel, such as the product owner. As a user, you can't enable or disable features like moderation within the application. Presently, in the development environment, the moderation flag is enabled. Future plans may include an option within the application for easier flag management.
{% endhint %}

<figure><img src="/files/18GVkvtyTRs3rXKZ6QvY" alt=""><figcaption></figcaption></figure>

Conversation of the bot and the showcase of LLM-Moderation **& Safety Layer**&#x20;

<figure><img src="/files/JPrjHzLiiaYYKuEO6XyN" alt="" width="563"><figcaption></figcaption></figure>


# Bot Governance

### **Overview**

As part of the **Security Architecture Review (SAR)** recommendations, organizations need to strengthen governance and observability for bots within the Interaction Experience (iX) ecosystem. This ensures that bot creation, deployment, and usage adhere to security, compliance, and operational standards.

### **Why Observability Matters**

* **Clear Visibility:** Know exactly which bots are being created, where they’re running, and whether they’re accessible to the public.
* **Managing Risks:** Stop unauthorized or insecure bots before they cause problems like data leaks or compliance breaches.
* **Strong Governance:** Make sure every bot follows your organization’s policies and security standards throughout its lifecycle.

### User Roles and Permissions

The **Bot Governance Dashboard** is designed with role-based access to ensure security and proper oversight. Below are the key roles and their respective permissions:

1. **Organization Admins**
   1. **Primary Audience**: The dashboard is mainly intended for organization admins.
   2. **Access Level**: Full visibility into all bots across the organization.
2. **Platform Admins**
   1. **Secondary Audience**: Platform admins also have access to the dashboard.
   2. **Access Level**: Similar visibility as org admins but primarily for operational support.

### How to access the Observability Report

1. This dashboard provides **visibility and governance** over all bots (or apps) created within the organization. It helps administrators track bot configurations, security posture, and compliance status.
2. Navigate to Administration-> Org Configuration-> Governance

<figure><img src="/files/f6SYB1CUD7VTOjuSPJK2" alt=""><figcaption></figcaption></figure>

3. The main section lists all bots created within the organization.
4. Columns include

| Settings        | Value                                                                           |
| --------------- | ------------------------------------------------------------------------------- |
| Name            | The bot/application name                                                        |
| IsPublicAccess  | Indicates whether the bot is publicly accessible (`true` or `false`)            |
| EnableAPIAccess | Shows if API access is enabled for the bot                                      |
| PIIEnabled      | Specifies whether Personally Identifiable Information (PII) handling is enabled |
| Created On      | Timestamp of when the bot was created.                                          |
| Action          | Provides options for further actions (e.g., view details, manage settings)      |

5. If the Governance team wants to understand more about the each of the apps, click on three dots and click on View App Details:

<figure><img src="/files/92Tp6PIgbBb4wgKJbuUn" alt=""><figcaption></figcaption></figure>

6. **App Details** view for a specific bot within the governance platform. This page provides comprehensive information about the selected bot, including its configuration, usage metrics, and associated components.
7. Summary Metrics

| Settings            | Value                                                                                               |
| ------------------- | --------------------------------------------------------------------------------------------------- |
| No of Publishing    | Total number of publishing instances for this bot (e.g., `3`)                                       |
| AI Contents Used    | Indicates the count of AI-generated content utilized (`0` here).                                    |
| Integrations in Use | Lists active integrations (e.g., `msgx4js.genAIs`).                                                 |
| Methods in Use      | Shows methods currently implemented (e.g., `getSpendDataList`, `chatcompletion`, `getPromoDetails`) |

<figure><img src="/files/TVAmc9OxIw8OtIiSYB1F" alt=""><figcaption></figcaption></figure>

When clicked on "Go to App" you will be redirected to the Custom App Publishing screen of the app "Acme Digital Assistant", which shows the Total number of publishing instances for this bot.

<figure><img src="/files/skSjZwx4XmBSLhY5rGSL" alt=""><figcaption></figcaption></figure>

### Enhance Bot Observability Dashboard on User Management

1. Governance Page can be accessed through User Management menu.
2. Navigate to User Management-> Manage User-> Org and User Management
3. Click the "View Details" button next to the organization "Concentrix".

<figure><img src="/files/VqIsLxoSBMvihsbzH1oM" alt=""><figcaption></figcaption></figure>

4. Selecting "View Details" lets users see a list of individuals associated with "Concentrix" and the apps they have created.

<figure><img src="/files/Jw6o5b02C54OPub46d56" alt=""><figcaption></figcaption></figure>

5. If the Governance team wants to understand more about the each of the apps, click on three dots and click on View App Details:

*Note: If you find apps with identical names, you can identify the correct one by checking the creator's username.*

6. Governance team members can access app details as described above:

<figure><img src="/files/B4masehh57h4nglPZKje" alt=""><figcaption></figcaption></figure>

7. When clicked on "Go to App" you will be redirected to the Custom App Publishing screen of the app "Acme Digital Assistant", which shows the Total number of publishing instances for this bot.

<figure><img src="/files/zcTzIe2Ogjx1XDg4ytXU" alt=""><figcaption></figcaption></figure>

This section explains how bot governance works, providing insights into bot creation, management, and processes. If necessary, governance can contact the bot creator to request modifications or removal. Currently, we have developed a report to offer more control and observability over bots.


# Manual Approval Governance for Public Sharing & Call/Text Enablement

### Overview

The Manual Approval Governance feature introduces an approval-based workflow for sensitive bot actions within an organization. It enables organizations to enforce administrative review and approval before specific actions are executed, helping improve security, compliance, and governance.

When enabled, requests related to public bot sharing, public web chat access, and Call/Text activation must be reviewed and approved by an Organization Administrator before they take effect.

***

## Why This Feature Matters

Previously, users could immediately perform the following actions:

* Share a bot with the entire organization
* Enable public access to a web chat link
* Activate Call/Text capabilities

These actions were executed instantly without administrative oversight, which could increase the risk of:

* Unintentional exposure of organizational bots
* Unauthorized public access to chatbot interactions
* Activation of communication channels without proper review

With Manual Approval Governance, organizations gain greater control over these actions through a structured approval process.

***

## Supported Approval Scenarios

### 1. Share Bot with Organization

Bot owners can request to share a bot across the organization.

After approval:

* The bot becomes available to all users within the organization.
* Authorized users can access and interact with the bot based on existing permissions.

<figure><img src="/files/co078HdHASK6t9Ba7XU7" alt=""><figcaption></figcaption></figure>

***

### 2. Enable Public Access to a Web Chat Link

Bot owners can request public access for a web chat link.

After approval:

* Anyone with the web chat URL can interact with the bot.
* Access is no longer restricted to organizational users.

<figure><img src="/files/gbYHxLlwIoJhFCf904dY" alt="" width="563"><figcaption></figcaption></figure>

***

### 3. Enable Call/Text Capabilities

Bot owners can request activation of Call/Text functionality.

After approval:

* Call/Text capabilities are activated.
* The feature follows the standard Call/Text activation process.

<figure><img src="/files/CtGHkMCwPA1b7oUzvgAc" alt="" width="563"><figcaption></figcaption></figure>

***

## Governance Configuration

Organizations can decide whether approval is required for specific actions.

### Available Settings

* **Mandate Manual Approval for Public Sharing**
* **Mandate Manual Approval for Call/Text Enablement**

#### Configuration Behavior

These settings are:

* Configured independently
* Disabled by default
* Applied only to new requests after activation

***

## Access Control

### Governance Configuration

**Location:**\
**Administration → Org Configuration → Governance**

#### Access Permissions

| Role            | Access      |
| --------------- | ----------- |
| Org Admin       | Full Access |
| Non-Admin Users | No Access   |

<figure><img src="/files/u5icqiG1NQtfBkWWJHiE" alt=""><figcaption></figcaption></figure>

***

### Governance Approval Queue

**Location:**\
**Administration → Governance**

#### Access Permissions

| Role                 | Access           |
| -------------------- | ---------------- |
| Org Admin            | Full Access      |
| Other Eligible Roles | Read-Only Access |
| Basic Role           | No Access        |
| Employee Role        | No Access        |

#### Approval Permissions

Only Org Admins can:

* Approve requests
* Reject requests

Other eligible users can:

* View approval requests

They cannot:

* Approve requests
* Reject requests

***

## Request Submission Process

When governance approval is enabled and a user initiates a controlled action:

#### Step 1: User Initiates Action

The user attempts to:

* Share a bot publicly
* Enable public web chat access
* Enable Call/Text functionality

#### Step 2: Confirmation Dialog

A confirmation message is displayed requesting user verification.

Share a bot publicly:

<figure><img src="/files/vNj9h3sc5yBCtxMTKyzf" alt="" width="563"><figcaption></figcaption></figure>

Approval for Web-chat:

<figure><img src="/files/E0qQr7uUEkPg0FB9S1G1" alt="" width="563"><figcaption></figcaption></figure>

Approval for call/text functionality:

<figure><img src="/files/wO5T0brllvy01c5W1GSg" alt=""><figcaption></figcaption></figure>

#### Step 3: Approval Request Creation

After confirmation:

* An approval request is generated.
* The request is added to the Governance Approval Queue.
* The request status is set to **Pending**.

<figure><img src="/files/s4tSr31nF6D43BHCTaWR" alt=""><figcaption></figcaption></figure>

#### Step 4: User Notification

The user receives confirmation that the request has been submitted for administrator review.

***

### When Governance Is Disabled

If approval governance is not enabled:

* No approval request is created.
* The action is executed immediately.
* Existing platform behavior remains unchanged.

<figure><img src="/files/BdMX79jd4nZSNNXUIbFn" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/dNyKu8VlIqXKl5h5uOe8" alt=""><figcaption></figcaption></figure>

***

## Governance Approval Queue

The Approval Queue serves as a centralized workspace for reviewing and managing governance requests.

### Information Displayed

Each request includes:

* Bot Name
* Bot Owner
* Requested Action Type
* Date and Time Submitted
* Current Status
* Request History
* Review Request Action (for Org Admins)

<figure><img src="/files/ay47CUr3X5wmezNuOxgH" alt=""><figcaption></figcaption></figure>

***

## Reviewing a Request

Org Admins can open any pending request to review additional details.

### Request Details

The Review Request dialog displays:

* Bot Name
* Created By
* Generative AI-Status
* Created Date
* Last Updated Date
* API Access Status
* PII Configuration Status
* Request Reason
* Approval Comments

***

### Available Actions

Org Admins can:

* **Approve**
* **Reject**

A confirmation step is required before the selected action is finalized.

<figure><img src="/files/y8hC4Sd7RNiZ5GRAISCW" alt=""><figcaption></figcaption></figure>

***

## Approval Queue Features

### Approval Type Filters

Users can filter requests using the following options:

* Pending - All (Default)
* Pending - Public Sharing
* Pending - Call/Text Enablement
* All Requests

<figure><img src="/files/mKdk1hgVJQM2bx9XKC17" alt=""><figcaption></figcaption></figure>

***

### Search

Search functionality allows users to quickly locate requests without manually browsing through the queue.

<figure><img src="/files/xMFfvksEjJAvQC8VLFz3" alt=""><figcaption></figcaption></figure>

***

### Refresh

The Refresh option reloads queue data without requiring a page refresh.

<figure><img src="/files/G7GKmPq6KE6FiiM21jyT" alt=""><figcaption></figcaption></figure>

***

### Request History

Every request includes a complete audit trail showing:

* Status changes
* User actions
* Administrative decisions
* Associated comments

This ensures transparency and accountability throughout the approval lifecycle.

<figure><img src="/files/kVWHLGFo4l6rSP1tg6Ko" alt=""><figcaption></figcaption></figure>

***

## Request Statuses

### Pending

The request has been submitted and is awaiting administrator review.

<figure><img src="/files/c91dE7IO6L0AUAcZ48i8" alt=""><figcaption></figcaption></figure>

***

### Approved

The request has been approved, and the requested action has been executed.

<figure><img src="/files/R8fgQ0WIDAaK4R9aEJoT" alt=""><figcaption></figcaption></figure>

***

### Rejected

The request has been denied and no action is performed.

<figure><img src="/files/COVHnr8e8ZrfuyWJFxm0" alt=""><figcaption></figcaption></figure>

***

### Cancelled

The request has been withdrawn or automatically cancelled by the system under specific conditions.

<figure><img src="/files/ODqpWyKdu2yPPZrgrygZ" alt=""><figcaption></figcaption></figure>

***

## Important Business Rules

### Audit Tracking

When a request is approved or rejected:

* Administrator comments are stored.
* The approving or rejecting administrator is recorded.
* Complete audit history is maintained.

***

### Duplicate Request Protection

The platform prevents duplicate approval decisions and safely handles concurrent review actions.

***

### Automatic Approval Scenario

If a Call/Text Enablement request is pending and the organization later disables the approval requirement for Call/Text governance:

* The pending request is automatically approved.
* System-generated remarks are recorded in the audit history.

***

## Multilingual Support

All governance-related screens support multilingual localization, including:

* Governance Configuration
* Governance Approval Queue
* Table Headers
* Filters
* Buttons
* Labels
* Dialogs
* Confirmation Messages
* Review Request Details

This provides a consistent user experience across all supported languages.

Example: Spanish Language

<figure><img src="/files/g7HoU2KLMTtz8pL0nD7L" alt=""><figcaption></figcaption></figure>

***

## Business Benefits

The Manual Approval Governance feature helps organizations:

* Control sensitive bot actions before execution.
* Reduce the risk of unintended public exposure.
* Ensure appropriate review of Call/Text enablement requests.
* Strengthen organizational governance and security controls.
* Maintain complete audit trails for compliance requirements.
* Provide a centralized approval management experience.
* Improve accountability and decision tracking.
* Support global teams with multilingual capabilities.

***

## Summary

Manual Approval Governance provides organizations with enhanced control over sensitive bot actions by introducing an administrator approval workflow. By requiring review before public sharing and Call/Text activation, organizations can improve security, maintain compliance, reduce operational risk, and ensure that changes are made with the appropriate level of oversight. This centralized approval framework also delivers full auditability, transparency, and multilingual support, making it suitable for organizations with diverse governance and compliance requirements.


# Manage Bot Visibility

This section explains the restrictions applied to bots using classified or private connections, designed to enhance security and comply with Security Assessment and Recommendations (SAR) and HRIX processes.

1. Feature Overview:
   1. Bots using private connections have restrictions on public access and conversation history.
   2. Alerts are provided to inform users on UI, when certain features are disabled due to private integrations.
2. Purpose:
   1. These restrictions aim to mitigate the risk of unauthorized data exposure, also known as bot exfiltration.
3. Management:
   1. Governance of private connections involves internal oversight to ensure security compliance.
   2. Infrastructure management is responsible for maintaining these restrictions.
4. Current Scope:
   1. These measures apply to internal users only but may be adapted for external requirements as needed.
5. Prerequisites :
   1. The bot must be integrated with classified or private connections recognized by the system.
   2. The system infrastructure must be configured to identify and manage these private connections.
   3. Users must have access to the latest system updates to ensure compliance with these restrictions.
   4. Applicable for any organization which has enhanced security requirements and want to restrict bot visibility for apps that have private data connection.

### Impact&#x20;

1. If any assistant is using a private connection, option to “grant public access” will be disabled, along with the reason and name of private connection.

   1. Click on “manage” for any available assistant:

   <figure><img src="/files/E8aRIPSoXAnKYdgqG2Fh" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/XTXARs8Qg1YjqUeSnpB1" alt=""><figcaption></figcaption></figure>

b. Option enabled in assistant settings when private connection is not used:

<figure><img src="/files/dssoJvd539MEgLhlkyh0" alt=""><figcaption></figcaption></figure>

2. Any assistant, if using a private connection, will be restricted from enabling call/text option with appropriate error message:

<figure><img src="/files/Hp5nTviLwpmFAdQTlSoH" alt=""><figcaption></figcaption></figure>

&#x20;     a. Error message displayed when user clicks on “Activate”:

<figure><img src="/files/u5ja0Mxc8vG9qM8yaDkm" alt=""><figcaption></figcaption></figure>

3\. For assistants with private connections, the conversation history will not be displayed or stored.

a. Open “chat history” for assistant with private connection:

<figure><img src="/files/i6NSJLQybvAtlqhbctF1" alt=""><figcaption></figcaption></figure>

b. For the selected bot, click on “submit”:

<figure><img src="/files/8AiJiDypjwqO4oPayHn1" alt=""><figcaption></figcaption></figure>

c. Error messages will be displayed:

<figure><img src="/files/L9i7T3vsotAzdW3WZFdu" alt=""><figcaption></figcaption></figure>

4. In Studio mode, Lex publishing is restricted for assistants with private connections.

<figure><img src="/files/jNT7TbmZDc7DlzOwBLF6" alt=""><figcaption></figcaption></figure>

a. Open assistant with private connection under custom apps:

<figure><img src="/files/QGsiL4pnfZDa4p7FVnG3" alt=""><figcaption></figcaption></figure>

b. Click on “Add app publishing” with Lex channel:

<figure><img src="/files/OCPguo55By6tYQ3kXk7n" alt=""><figcaption></figcaption></figure>

c. Save button will be disabled with reason and name of private connection mentioned:

<figure><img src="/files/BBl6N17cL4744GC7RLN3" alt=""><figcaption></figcaption></figure>


# Shared Bot Access Management

### Published Bots

We have a new feature called **Published Bots**, which serves as a **centralized dashboard** to view all bots available to a user. This includes:

* Bots created by the user
* Bots shared directly with the user
* Bots shared at the organization level

This new experience significantly simplifies bot discovery and interaction.

<figure><img src="/files/oJMdnULZMrAw1rLEEbGS" alt=""><figcaption></figcaption></figure>

Published Bots provides a comprehensive overview of each bot, including:

* **Custom App Name**: Since one app can host multiple bots.
* **Published Bot Name**
* **Phone Number**: If assigned.
* **GenAI Usage**: Indicates if the bot uses GenAI.
* **Sharing Status**: Specifies if the bot is private, shared, or available organization-wide.

This summary gives users a complete view of all bots accessible to them.

Users can **chat directly with any published bot** from the Published Bots page. The chat functionality is identical to using the bot within its original application. Note that you can only chat here—**no editing or bot management allowed**. This feature enables quick interactions with shared or organizational bots.

<figure><img src="/files/K27QHb5uZnUzkm9qh7Q9" alt=""><figcaption></figcaption></figure>

The menu includes options such as:

* **Created by me** – bots the user owns
* **Shared with me** – bots shared by others
* **All** – a complete list of accessible bots

Based on your choice you can filter the bots available.

<figure><img src="/files/9T6DiLirneib7yRSsMe0" alt=""><figcaption></figcaption></figure>

### Updates to Custom Apps

Previously, users could view all shared apps here, but now it is restricted to **Created by me**. This is because published/shared bots are now surfaced under the new **Published Bots** section.

The idea is to clearly separate:

* **Editable bots** → stay under **Custom Apps**
  * These are bots the user owns and can fully customize
* **Published/shared bots** → moved to **Published Bots**
  * These offer a chat-only interface and cannot be edited

<figure><img src="/files/2CTqp21vL8REZX2ZhKU8" alt=""><figcaption></figcaption></figure>

In the future, beyond org-level sharing, we will also support **team-level sharing**, and these shared bots will continue to appear under Published Bots.


# Controlled Bot Sharing & Team Management

## Overview

In iX Hello we have two interrelated features:

* Bot sharing capability - Enables  sharing bot  at the Organization or Teams Level.
* Team Structure Management - Involves syncing team data from the Admin Center and populating it within iXHC.

## Team Structure Management

Team Structure Management in iX Hello Customer lets organizations create, organize, and manage teams of users within the platform. Teams provide a structured way to support collaboration, access control, governance, and resource sharing, including bots, apps, and data.

## Teams based bot sharing in iX HC

To enable more controlled and flexible sharing, iX Hello Customer now support bot sharing with Team and Organization.

Previously, custom apps could be shared across the entire organization, which unintentionally granted all users access to view and modify bot configurations. This approach has now been deprecated to enhance security and governance.

With the introduction of the Share Bot feature:

* Users can share bots with specific teams or individuals within the organization.
* Shared users gain access only to the bot’s chat functionality via the Published Bots screen.
* Configuration settings remain protected, ensuring that only authorized owners or contributors can make changes.

This enhancement ensures that:

* Bots are visible only to relevant users or teams.
* Configuration integrity is maintained.
* Organizations can still scale sharing when needed, without compromising control.

#### **What You Can Do**

* Choose how to share your bots
  * Keep them private
  * Share with the entire organization
  * Or share only with specific teams
* Collaborate easily
  * Work closely with selected teams by giving them access to the bots they need
* Control who can see your bots
  * Ensure bots are visible only to the right audience, avoiding unnecessary exposure
* Hassle-free sharing
  * In the current phase, all sharing requests are automatically approved, ensuring a seamless and frictionless collaboration experience.

    In future releases, an approval workflow will be introduced to provide an additional layer of security and governance. This enhancement will enable organizations to review and control access requests more effectively, aligning with stricter access management policies.

#### Feature Availability

The Share Bot option is available:

* In Studio Mode
  * Custom Apps (during publish)
  * Quick Start Apps

Note: The sharing option is available after bot creation, not during the creation process.

#### 1. Organization/Teams based bot sharing in Studio Mode

In Studio Mode, the Share Bot option enables bot creators to control how their bots are accessed by others either across the organization or limited to specific teams.

This is especially useful in Studio Mode, where bots are typically built and deployed as part of Custom Apps or Quick Start Apps.

**Where the Share Option Appears**

In Studio Mode, the Share Bot option is available during the Publish stage:

* Custom Apps → Publish
* Quick Start Apps&#x20;

After creating a bot, navigate to the Publish tab and click “Share” from the bot actions menu or detail page.

* A “Share External App” popup will be displayed

Studio Mode integrates sharing into the deployment workflow, ensuring bots are shared only when they are ready for use.

<figure><img src="/files/a40hoq8J094ASeE8rXKk" alt=""><figcaption></figcaption></figure>

When the user selects “Share Bot”, the Share External App modal presents the following options:

<figure><img src="/files/pNRqbUAQIh3wB9WxHouV" alt=""><figcaption></figcaption></figure>

1\. Keep Private

* The bot remains accessible only to the owner
* It is not visible to any other users

2\. Share Across Organization&#x20;

* The bot becomes available to all users within the organization

3\. Share Across Team

* The bot is shared only with selected teams that the owner belongs to
* This enables targeted access, ensuring the bot is available only to relevant groups instead of the entire organization
* Helps maintain better access control and focused collaboration

**Share Bot request Submission**

* The Share button is enabled only after selecting at least one option either organization level or team level sharing (at least one team should be selected)
* On submission:
  * A sharing request is created
  * Sent to the approval queue
  * Currently, requests are auto approved

**Let us focus on Team-Based Sharing Flow**

When “Share Across Team” is selected:

Share across team is disabled when the user is not part of any team. Therefore, the system shows:*“You are not a member of any team.”*&#x20;

“Team Creation Guidelines” link is provided to help the user get started.

When you click on the link, you will be navigated to a Guidelines page, that helps users understand how to enable the “Share Across Team” feature by first creating or joining teams within the platform.

<figure><img src="/files/njf8qf2dzxQgd0wWsYNN" alt=""><figcaption></figcaption></figure>

Once you land on the Admin Center page, you will be able to select the team or add the team by clicking on "Add Team"

<figure><img src="/files/jK4onFVsr5AcvBWYbcAk" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/vHWHebuFlhqkpMOj9crK" alt=""><figcaption></figcaption></figure>

Once after you created the team, add members to the team, by clicking on "Add Team Members"

<figure><img src="/files/y6B5mvRC3JrlWIOB8jeQ" alt=""><figcaption></figcaption></figure>

Once the member is added to the Team, you will see a success message and also the list of team members will be visible under the team which you created.

<figure><img src="/files/PdnS0XOBfswaUgORxO87" alt=""><figcaption></figcaption></figure>

**Teams Data Synchronization Overview**

The iX Hello Customer platform automatically syncs team data from the iXHello Employee Admin Center to support team-based bot sharing and access control.

* The platform receives team data change notifications through secure webhook events.
* These events are processed to maintain an up-to-date local copy of team structures, including team hierarchy and membership.
* Built-in safeguards ensure secure processing, data accuracy, and reliability.

**Outcome**

* Teams from the Admin Center are automatically synchronized with the iX Hello Customer platform.
* These teams can be used to share bots with specific teams.
* Bot access is controlled using latest synced team data.

As a result, users can confidently share bots within teams, knowing that access is always based on current and accurate team information.

**Team Selection**

* The system fetches all teams the bot owner belongs to
* Displays them as a selectable checklist
* User can select one or more teams
* If no teams exist: “You are not a member of any team.”

**Approval Handling**

* Approval is automatic in the current phase
* The system processes and marks the request as Approved

**Post Approval Behavior**

Once approved:

* The bot becomes available within selected team
* Team members can:
  * Discover the bot
  * They can access the chat feature through the "Published Bot" menu.
  * Only Bot owners has the authority to modify the bot settings.

<figure><img src="/files/ZTk0D9VuqQ46aMrSIO3B" alt="" width="563"><figcaption></figcaption></figure>

Click 'Apply' to share the bot with the selected team.&#x20;

<figure><img src="/files/nvBARXm1lACsCAHjPiif" alt=""><figcaption></figcaption></figure>

#### Teams based bot sharing in Quickstart

If you need more information on how quick start apps works refer to [this](/ixhc/general/rag-file-upload/file-upload-in-studio-mode-1/quick-start-apps).

1. Accessing Quick Start Apps

* Navigate to Quick Start App from the platform.
* This will open the Quick Start Apps dashboard.
* Under “My Assistants”, you will find all the bots you have created using Quick Start.

2. Sharing a Quick Start App

* Select an existing app that you want to share, or create a new bot.
* Click on the three dot menu (⋯) available on the bot.
* From the options, select “Share”.

<figure><img src="/files/fG44a0MOjdnXv6tJ4eQc" alt="" width="563"><figcaption></figcaption></figure>

3. Share Modal Opens

* A popup titled “Share External App” appears.
* This is the same sharing interface used across Studio Mode, ensuring a consistent experience.

4. Select “Share Across Team”

* Choose the “Share Across Team” option.

This allows you to share the bot in a controlled manner, instead of making it available to the entire organization.

5. Select Teams

* The system automatically fetches all the teams you are a part of.
* These teams are displayed as a selectable checklist.
* You can choose one or more teams to share the bot with.

<figure><img src="/files/pjNrpthrHXo0RTtOkZIB" alt="" width="563"><figcaption></figcaption></figure>

If you are not part of any team, then follow the above steps mentioned on the Studio mode - Teams based bot sharing.


# RAG File Upload

Retrieval-Augmented Generation (RAG) with file uploads, you'll typically need to integrate a file upload mechanism with a RAG pipeline that processes and indexes the uploaded data. This allows users to query the uploaded files using natural language.&#x20;

**Custom AI Assistants and Document Processing**

Our platform enables the development of custom AI assistants that can efficiently process and utilize knowledge from a variety of uploaded documents stored locally. This feature enhances both data security and accessibility, ensuring that your information is protected while being readily available for AI processing when needed.

The platform supports a wide range of file formats, including:

* **PDF**: Widely used for documents due to its consistent formatting across different devices.
* **DOCX**: Commonly used for text documents with complex formatting, originating from Microsoft Word.
* **TXT**: A simple format for text files without any special formatting, making it versatile for basic document storage.
* **CSV**: Used for storing data in a comma-separated values format, perfect for spreadsheets and databases.

By integrating seamlessly with your local storage, our AI assistants can access and interpret these formats to provide enhanced services and insights, while also ensuring that your data remains within your control. This blend of functionality and security makes our platform a reliable choice for organizations looking to leverage AI in processing diverse document types.

1. **File Upload:**\
   Users can upload files in PDF, DOCX, TXT, or CSV formats through a local storage interface.

* **Validation:**\
  The system checks each file's format and size to ensure they meet the specified limits.
* Increased training file size from 1MB to 5MB per document
* Increased number of upload-able files from 10 to 20 files
* **Indexing:**\
  Files are indexed and divided into sections for easy retrieval within a knowledge base designed for Retrieval-Augmented Generation (RAG).

2. **Training & Indexing:**

* Extracts, preprocesses, and indexes document contents.
* Offers configurable chunking strategies (e.g., semantic, fixed size).
* Enhances retrieval context through metadata extraction.
* OCR Document scraping training functionality

3. **Query & Retrieval:**

* Retrieves relevant document snippets during assistant interactions.
* Ensures AI can utilize uploaded knowledge for accurate responses.

4. **UI/UX Requirements:**

* Users can view, manage, and delete uploaded files.
* Allows file refreshing or reprocessing as needed.

5. **Security & Compliance:**

* Access controls limit file uploads to authorized users.
* Encryption and local storage policies meet CNX App Sec compliance standards.

### File Upload Options in iX Hello

iX Hello provides two modes for using the File Upload options:

* **Basic Mode**: This mode is designed for users who prefer a straightforward, easy-to-use interface. It offers essential features for quick file uploads without additional configuration or complexity.
* **Studio Mode**: This mode is suitable for advanced users who require more control and customization over the file upload process. It includes enhanced features and settings that allow for more detailed adjustments.


# File Upload in Basic Mode

## Pre-requisite:

a.     You need to have an iX Hello account for this. Contact our team to have your account created.

b.     Login in and Click on Add a New Assistant.

This takes you to iX Hello Blueprint Library, which is a collection of templates that you can use to create ready-to-use AI-Apps. Refer to [this](/ixhc/general/ix-hello-basic-mode) to create an app in Basic Mode.

You can explore the possibilities here. But for now, let’s focus on creating the GPT-Service Assistant. On the search bar, type ‘GPT-Service Assistant’.

<figure><img src="/files/MjigQCFIDTTszgX4k4EU" alt=""><figcaption></figcaption></figure>

Click on Create App. Fill the details as you proceed.

Now let's concentrate on File Upload:

When you click on the training tab after filling up all the other details,

You get to see a feature called "Upload Files", this feature supports multiple file formats and integrates with local storage, ensuring data security and accessibility.

**File Formats**

* .PDF
* .CSV
* .DOCX
* TXT

The "Choose Files" option allows you to upload data from your local device or computer to an online server or storage system. This feature facilitates the sharing of various file types, such as documents, images, and videos, over the internet.

Once file is uploaded then click on Save button.

<figure><img src="/files/laMFc5411OcVtGpzzm88" alt=""><figcaption></figcaption></figure>

When creating an Assistant, users can upload up to 10 files, with each file not exceeding 1 MB in size.

When a user attempts to upload more files than allowed, the application will display an error message indicating the limit has been exceeded.

<figure><img src="/files/qbBF0wj6DftYp8OG5mpm" alt=""><figcaption></figcaption></figure>

Once the files are uploaded successfully you get see the below screen.

<figure><img src="/files/8fi1vMnHXEUP7IaoLZN3" alt=""><figcaption></figcaption></figure>

Here's an explanation of how the file upload feature operates in Basic Mode.


# File Upload in Basic Mode

## Pre-requisite:

a.     You need to have an iX Hello account for this. Contact our team to have your account created.

b.     Login in and Click on **Add a New Assistant**.

This takes you to the iX Hello Blueprint Library, which is a collection of templates that you can use to create ready-to-use AI-Apps. Refer to [this](/ixhc/general/ix-hello-basic-mode) to create an app in Basic Mode.

You can explore the possibilities here. But for now, let’s focus on creating the GPT-Service Assistant. On the search bar, type ‘GPT-Service Assistant’.

<figure><img src="/files/MjCN8HbFKXI4fRbZFVuX" alt=""><figcaption></figcaption></figure>

Click on **Create App**. Fill the details as you proceed.

### To upload files:

After filling up all the other details, open the Training tab.

Scroll down until you see a feature called "Upload Files". This feature supports multiple file formats and integrates with local storage, ensuring data security and accessibility.

**Accepted File Formats**

* .PDF
* .CSV
* .DOCX
* .XLSX

The "Choose Files" option allows you to upload data from your local device or computer to an online server or storage system. This feature facilitates the sharing of various file types, such as documents, images, and videos, over the internet.

<figure><img src="/files/AXnz0Q9Gv9zCdb9pqNUn" alt=""><figcaption></figcaption></figure>

When creating an Assistant, users can upload up to 10 files, with each file not exceeding 1 MB in size. When a user attempts to upload more files than allowed, the application will display an error message indicating the limit has been exceeded.

<figure><img src="/files/wLimauqiFFk4CJgxqkxz" alt=""><figcaption></figcaption></figure>

Once the desired files are uploaded, click on **Save** button. Once the files are uploaded successfully you get see the below screen.

<figure><img src="/files/8fi1vMnHXEUP7IaoLZN3" alt=""><figcaption></figcaption></figure>

The files are now available for use by the assistant.


# File Upload in Studio Mode

The purpose of this document is to describe how to create a generative AI app in the studio mode in iX Hello. After completing the following series of documents, you should be able to create a generative AI app for web chat and phone/text. This document also explain on how the File Upload works in Studio Mode.

a. [Prerequisites](/ixhc/general/ix-hello-studio-mode/prerequisites)

b. [Steps to Build Custom App](/ixhc/general/ix-hello-studio-mode/steps)

Let's have a look at how File Upload works in Studio Mode:

In AI-Contents  you see an option called Add New AI Content.

Click on Add New AI Content.

<figure><img src="/files/tNSgOOmgTNF7L6lNOhUJ" alt=""><figcaption></figcaption></figure>

Once clicked on Add New AI content, user will be navigated to the below screen:

<figure><img src="/files/h2x5qPcOBtQ1GaWPWOoZ" alt=""><figcaption></figcaption></figure>

**Add Training Content and URLs**

You can directly add Training URLS, and for text content, you can select from available list of text content from the drop-down menu and click on Save.

**Upload Files:**

This feature supports multiple files formats and integrates with local storage, ensuring data security and accessibility.

**File Formats**

* .PDF
* .CSV
* .DOCX
* TXT

The "Choose Files" option allows you to upload data from your local device or computer to an online server or storage system. This feature facilitates the sharing of various file types, such as documents, images, and videos, over the internet.

Once file is uploaded then click on Save button.

Users can upload up to 10 files, with each file not exceeding 1 MB in size.

When a user attempts to upload more files than allowed, the application will display an error message indicating the limit has been exceeded.

<figure><img src="/files/rP2qfx04cminYlYMFMCX" alt=""><figcaption></figcaption></figure>

After the files are uploaded, you can view them listed under "Files Selected." These chosen files will serve as the training data for the bot. This means that before responding to any user's queries, the bot will thoroughly analyze the attached documents and URLs provided. The comprehensiveness of this training data is vital as it equips the bot with the necessary context and information, ensuring more accurate and relevant responses to inquiries. Hence, ensuring the right files are selected for upload is crucial, as they form the foundation for the bot’s learning and subsequent interaction quality.

<figure><img src="/files/LvHXeKzHLpDQ18XnHTpO" alt=""><figcaption></figcaption></figure>


# File Upload in Studio Mode

The purpose of this document is to describe how to create a generative AI app in Studio Mode of iX Hello. After completing the following series of documents, you should be able to create a generative AI app for web chat and phone/text. This document also explain on how File Upload works in Studio Mode.

a. [Prerequisites](/ixhc/general/ix-hello-studio-mode/prerequisites)

b. [Steps to Build Custom App](/ixhc/general/ix-hello-studio-mode/steps)

Let's have a look at how File Upload works in Studio Mode:

In AI-Contents, you see an option called Add New AI Content.

Click on **Add New AI Content**.

<figure><img src="/files/4O7QNTrUg4ZACEYkEzvK" alt=""><figcaption></figcaption></figure>

The user will be navigated to the following screen:

<figure><img src="/files/h2x5qPcOBtQ1GaWPWOoZ" alt=""><figcaption></figcaption></figure>

**Add Training Content and URLs**

You can directly add Training URLS, and for text content, you can select from available list of text content from the drop-down menu and click on Save.

**Upload Files:**

This feature supports multiple files formats and integrates with local storage, ensuring data security and accessibility.

**File Formats**

* .PDF
* .CSV
* .DOCX
* .XLSX

The "Choose Files" option allows you to upload data from your local device or computer to an online server or storage system. This feature facilitates the sharing of various file types, such as documents, images, and videos, over the internet.

Once file is uploaded then click on Save button.

Users can upload up to 10 files, with each file not exceeding 1 MB in size.

When a user attempts to upload more files than allowed, the application will display an error message indicating the limit has been exceeded.

<figure><img src="/files/oaQ4JWRzdyIWM0oHXLCJ" alt=""><figcaption></figcaption></figure>

After the files are uploaded, you can view them listed under "Files Selected." These chosen files will serve as the training data for the bot. This means that before responding to any user's queries, the bot will thoroughly analyze the attached documents and URLs provided. The comprehensiveness of this training data is vital as it equips the bot with the necessary context and information, ensuring more accurate and relevant responses to inquiries. Hence, ensuring the right files are selected for upload is crucial, as they form the foundation for the bot’s learning and subsequent interaction quality.

<figure><img src="/files/ju0yZt0RQdmBoR4Vcub7" alt=""><figcaption></figcaption></figure>


# Quick Start Apps

Quick Start Apps in **iX Hello** are like **ready-to-use blueprints** that help you quickly build conversational applications without starting from scratch. Instead of writing all the logic, queries, intents, and flows by yourself, you can pick one of these apps and then just **customize it** for your needs.&#x20;

They are designed for people who want to:

* Quickly connect iX Hello with databases and Faster deployment
* Integrate with ticketing tools or Sample data models
* Connect to CRM systems&#x20;
* Automate common ITSM and business use cases
* Ready-made integration flows
* Predefined intents & utterances

**Quick Start Apps**: This fixed button is accessible in Studio Mode and appears on all pages and menu items.

<figure><img src="/files/dKJ1kOoNntZIpxXLIPAG" alt=""><figcaption></figcaption></figure>

### Quick Start App: Building an application

To build an application using the Quick Start App, follow these steps:

1. Open the Quick Start App.
2. Fill in all required fields under the **General Information** section.
3. Navigate to the **Knowledge Base** tab.
4. Use the file upload feature to upload all necessary knowledge bases (KBs).
5. Click the **Save** button to store all entered information.

<figure><img src="/files/0tbidp4syBffsXxF6bV7" alt="" width="563"><figcaption></figcaption></figure>

<figure><img src="/files/S2D3HZF3OnKPI4cy4k4Q" alt="" width="400"><figcaption></figcaption></figure>

To create the app, click the **Save** button. Then, click **Activate** to start interacting with the bot.

<figure><img src="/files/eM3C3XWEMlPYOHOLCSNp" alt="" width="563"><figcaption></figcaption></figure>

This is how Quick Start Apps in iX Hello make it easy to get started quickly with ready-to-use chatbot solutions.”




---

[Next Page](/llms-full.txt/1)