# Aisera Product Documentation Welcome to the Aisera Product Documentation Portal! The Product Documentation includes step-by-step instructions that describe how to create bots, use APIs, and administer your instance of the Aisera Gen AI software platform. ### Documents:

	Cover image
Discover the applications you can build using the Aisera Gen AI Platform	/files/w2FNu8nYnUrrd6XpGvyO	https://docs.aisera.com/overview-of-aisera
Use the bot creation feature to create your own pre-trained assistant.	/files/1VHSMmfVuJoJ2gEi6Jrw	https://docs.aisera.com/aiseragpt
Add the Agent Assist bot to the applications used by your Support and Service teams so they can compare similar tickets to the one they're currently working on.	/files/W0DNn3J6IiCQEqEX1gKc	https://docs.aisera.com/agent-assist
Store the documents you use the most within your AI Copilot dashboard and personalize your customer contact experience.	/files/822eRcFfpPdNRoDWrcJ1	https://docs.aisera.com/aisera-copilot
Analyze the data for your enterprise ticket and knowledge information to track incidents and pro-actively find solutions.	/files/idQ0ee7yAr44bani2x8K	https://docs.aisera.com/ai-ops
The Aisera Ingestion APIs allow you to upload data to your Aisera vector database directly.	/files/w43wothScHvZPN1PxwMn	https://docs.aisera.com/apis
Customize your applications and bots using the Aisera Admin UI to configure settings, map fields, add your own transform scripts and create workflows.	/files/MFAf36abKdpPd3bR58r9	https://docs.aisera.com/aisera-platform
Read the Aisera Product Release Notes every two weeks to keep track of the latest features and changes in the Aisera Admin UI.	/files/HLgWdFh57md0bQCFdm3O	https://docs.aisera.com/product-release-notes
The self-service, AI agent builder.	/files/tJZZ5PmCqUD7LrwCd68X	https://docs.aisera.com/unify

Use the documentation to become familiar with the Aisera products, so you can work productively in collaboration with your Aisera team. Feel free to contact your Aisera team if you see errors, omissions, or changes that need to be made to this documentation. # Aisera Gen AI Platform Overview The Aisera product is a leading generative, [agentic AI](https://aisera.com/blog/agentic-ai/) platform created to transform enterprise service experiences through the integration of generative AI and advanced automation. It leverages Large Language Models (LLMs) and [domain-specific AI](https://aisera.com/blog/domain-specific-ai-agents/) capabilities to deliver proactive, personalized, and predictive solutions across various business functions such as IT, customer service, HR, and more. * If you're new to the Aisera Gen AI platform and you want to become familiar with terms and concepts, proceed to the [**Getting Started Guide**](/overview-of-aisera/getting-started-guide) after reading this topic. * If you're already familiar with the Aisera Gen AI platform, proceed to [**End-to-End Bot Set up Steps**](/aisera-platform#end-to-end-bot-set-up-steps) * [**Using the Product Documentation**](#using-the-product-documentation) #### Key Features and Capabilities **1. Generative AI Models and Studios:** Aisera's platform includes a suite of generative AI models and studios that facilitate the development of AI applications. This includes the LLM Studio for creating and refining prompts, the [Event Studio](https://aisera.com/blog/event-studio/) for dynamic action flows, and the Workflow Studio for automating complex business processes with a low-code/no-code interface. These tools help in building customized AI solutions efficiently and cost-effectively. **2. Conversational AI:** The platform's Conversational AI capabilities are designed to provide human-like interactions across multiple domains and languages. This includes features like intent matching, context switching, and disambiguation to ensure accurate and relevant responses. It supports self-service resolutions and integrates seamlessly with communication platforms like Slack and Microsoft Teams. **3. AI-Driven Service Management:** Aisera offers a [next-generation IT Service Management](https://aisera.com/products/next-gen-itsm/) (ITSM) solution that automates service requests, incident management, and change management. Its AI-native ticketing system is built to enhance operational efficiency by predicting and resolving issues proactively. This integration extends to various enterprise systems, improving the overall employee and customer experience. **4. Enhanced Customer and Employee Experiences:** Aisera's platform includes AI Copilot and [Agent Assist](https://aisera.com/products/assist/) features that provide real-time recommendations and next-best actions, improving productivity and satisfaction. These tools use AI to analyze past interactions and provide insights, ensuring that agents and employees can resolve issues quickly and effectively. **5. Advanced Analytics and Reporting:** The platform offers interactive analytics and user profile-based recommendations, allowing organizations to personalize service resolutions and gain insights into user behavior. This helps in making data-driven decisions and improving service quality continuously ## Using the Product Documentation This Aisera Product Documentation takes the form of a task-oriented description of the Aisera Admin UI, since you need to use the Admin UI to complete the tasks that you have to perform to build your application/bot functionality.\ \ When you first log into your Aisera tenant endpoint, you will see the Aisera Admin UI. This is the development tool that allows you to configure your tenant settings and create applications/bots to use within your Enterprise applications.

The Aisera Product Documentation describes how to use the Aisera Admin UI. After you have logged in to your Aisera tenant endpoint, and you can see the development tool, open a separate browser window and type `docs.aisera.com`. (You've probably already done this, if you're reading this.) Use the global Search field in the top right corner of the Aisera Product Documentation portal to find a topic within the product documentation set.

Select a heading at the top of the Product Documentation portal to see a drop-down list with each of the topics in this guide.

Expend the side navigation menu within each topic to see all of the related pages.

# Getting Started Guide If you're new to the Aisera Gen AI Platform, and you want some guidance about where to start, this section presents you with options to get to know the Aisera Admin UI and the bot-building process better.\ \ The steps involved include: 1. Becoming familiar with the **Aisera Gen AI Platform**, terms, and prerequisites. 2. Integrating your **External Data Source** (customer support software) with your Aisera Gen AI **Tenant**. 3. **Ingesting** data from your External Data Source to the Aisera Tenant cloud datastore. 4. Deciding what you want the application/bot to accomplish. 5. Building an **Application/Bot**. 6. Activating Initial Phrases. 7. Customizing Prompts (describe the bot personality and goals). 8. Adding **Workflows** or **Hyperflows** (what you want the bot to do on the backend). 9. **Test** the application/bot. 10. **Run** the application/bot and **Train** it using user feedback. 11. Review **Analytics** and **Metrics**, and optimize your application/bot as needed. The Aisera Gen AI Platform provides many ways to customize, test, and optimize your application/bot. Before you begin, see the **End-to-End** list. # Glossary of Aisera Terms Make sure you understand these [**general AI terms**](https://aisera.com/blog/ai-terms-you-need-to-know/) and the have read the [**Prerequisites**](https://docs.aisera.com/overview-of-aisera/getting-started-guide/prerequisites) section before you proceed to build your applications/bots. ## Aisera Products | Aisera Product | Description | | -------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Agent Assist |

Agent Assist is an embedded application/bot that runs in your Agent Desktop such as ServiceNow, Salesforce, Zendesk, or Jira Service Desk and provides IT Helpdesk Agents with recommendations to resolve their cases or real time live agent conversations occurring over chat or phone.

Currently supported on Zendesk, ServiceNow, and Salesforce Service Cloud and Jira Service Desk.
Includes Agent Assist Analytics display page within the Aisera Admin UI.
Terms not to use: Auto-resolve, TicketIQ, TicketAI, AI Assist
Modules include:
- Case Summary: Quick summary of what’s happened in the Ticket thus-far. This helps tier 2 / tier 3 escalation resources quickly catch-up on what’s happened thus-far.
- Aisera Answer: RAG answer. Typically a richer answer than an end-user receives as the service desk often has access to more granular technical documentation than what’s available to the general employee/customer audience.
- Next Best Actions: Recommended next-steps based on learnings from historical similar cases.
- Field Predictions: Predicts how to categorize, assign, route, or prioritize based on historical ticket learnings. Can auto-apply predictions above a certain threshold (ie. above 90% confidence in prediction).
- Intelligent Swarming: Aisera offers intelligent swarming through analyzing various factors like ticket content, urgency, impact, and historical data to identify tickets that will likely require collaboration from multiple experts across various departments
- User Insight/Sentiment Analysis: Aisera’s Sentiment analysis uses natural language processing (NLP) and machine learning techniques to analyze and interpret the emotions expressed in text data from an utterance or the conversation thread associated with a user query.
- Case Wrap-Up: Allows IT technicians, engineers, or Customer Support Agents to look up the Case History for future issues or escalations.
- Case Creation: The Aisera platform includes out-of-the-box workflows for automated ticket creation and ticket management. Aisera can create and update cases in any ticketing system from any conversational channel, such as Slack, Microsoft Teams, or a custom conversational interface.

Use the Ticket AI Agent Assist Channel. Add other channels, such as IVR as needed.

| | Aisera Assistant (formerly AI Copilot) |

Aisera Assistant is an advanced, enterprise-grade employee copilot designed to enhance productivity and streamline workflows through proactive automation and intelligent task management. Leveraging the power of AiseraGPT and domain-specific large language models (LLMs), Aisera Assistant delivers personalized, context-aware support that caters to a wide range of business needs.

With Aisera Assistant, users can upload documents locally and save previous conversations.

Terms to Use: Agent, Universal Agent, Agent Orchestrator, Conversational AI
Terms not to use: Bot, Copilot, AiseraGPT, App
ConversationalAI 2.0:
- Flow Search

See also: Overview of Aisera Assistant/Copilot

| | AI Ops Assistant/Agentic AI |

This application can be built today using the AI Ops base application with an AI Copilot channel. Some of the capabilities are still in development.
An AI system built to act as an autonomous agent(s). Capable of conversing, reasoning, and performing actions dynamically with minimal human intervention.

Key Components of an Agentic AI:
- Autonomous: operates independently based on goals or prompts
- Multi-Fullfillment / Task Planning: able to dynamically orchestrate a diverse set of tools / skills / sub-agents (Knowledge Serving / RAG, Code Gen, Workflows, form and execute API requests, text-to-sql, etc.).
  - Truly agentic agents should be able to invoke these tools sequentially and/or in parallel. They should be able to create a plan of actions and/or perform task decomposition. If the AI is limited to singular request/response tool calling, this is not agentic.
- Context-awareness: Ability to understand and maintain conversational context as well as adapt to environmental / user-specific context (environmental variability, user personalization / ACL adherence)
  - Agentic agents need the ability to store/cache not just conversation history but also output/environment variables from previously invoked tools / sub-agents
- Proactivity: Can initiate tasks, not just respond to commands.

See also: Security & Privacy (TRAPS).

| # Tenant Configuration You can set Tenant Configuration parameters when you are setting up your Aisera Gen AI Platform Tenant. {% embed url="" %} # Add Integration (login credentials) to Tenant {% embed url="" %} # Add Channel to Tenant {% embed url="" %} # Add Data Source to Tenant {% embed url="" %} # Agent Assist Window The Aisera Admin UI contains an analytics window specifically for the Agent Assist (called Aisera Assist in some places) application/bot. {% embed url="" %} # Analytics ### Pre-Built Analytics {% embed url="" %} # Application/Bot Use the videos in this section to view the way that actions are performed for the Application/Bot within the Aisera Admin UI. # Agent An AI Agent is an application/bot that is often used within an Enterprise application. It learns from existing **Tickets** or **Knowledge Base Articles** and returns answers when users type in questions. In this documentation, a human performing this task is known as a Service Agent. An AI performing these tasks is referred to as an application/bot. Within the Aisera Admin UI, an Agent looks like this while it is in development:

To view an existing bot using the Aisera Admin UI: And it looks like this when it is deployed:

The **Channel** associated with an application/bot determines what the deployed bot will look like. ### View an Existing Aisera Bot {% embed url="" %} ### Create a New Aisera Bot {% embed url="" %} Proceed to **Add a Data Source**. # Data Source ### Add a Data Source to an Aisera Bot {% embed url="" %} # Add/Register Workflow If you have workflows or hyperflows associated with your Tenant, you can choose them and associate them (also called Register them) with your Application/Bot. ### Add a Workflow {% embed url="" %} # AiseraGPT Policy Choose **View Policy** in your Bot Details window to see policies associated with your Application/Bot. {% embed url="" %} # Knowledge Generation Policy Select the View Knowledge Generation Policy button to see AI models associated with your Knowledge Article Generation. {% embed url="" %} # Bot Channel ### Default Channel {% embed url="" %} ### Add a Channel {% embed url="" %} ### Channel Details {% embed url="" %} # Add Live Agent {% embed url="" %} # Error Handling Flow {% embed url="" %} # View Conversation Messages {% embed url="" %} # View Intents {% embed url="" %} # Access Attributes {% embed url="" %} # Audit Trail Logs {% embed url="" %} # AI Automation {% embed url="" %} # AI Workbench The **AI Workbench** menu allows you to access the **Unresolved**, **Reviews**, **Health**, and **Test Suites** windows that help you monitor **Requests** from your application/bot. {% embed url="" %} See [**Requests**](https://docs.aisera.com/aisera-platform/ai-workbench-optimize-conversations/requests-window) for more information. # AI Discovery - CMDB {% embed url="" %} # AI Lens {% embed url="" %} # AI Observability {% embed url="" %} # AI Generated Documents {% embed url="" %} # Campaigns ### Create a Campaign {% embed url="" %} # Content Generation {% embed url="" %} # Event Studio ### Open Event Studio {% embed url="" %} ### Create an Event Trigger {% embed url="" %} # Hyperflows ### Create a Hyperflow {% embed url="" %} ### Add a Function To add a a **Function** to the **Hyperflow**, the Hyperflow must be in **Draft** model. # Knowledge Base Articles This publication uses KBA instead of the abbreviation KB to refer to Knowledge Base Articles because it also discusses knowledge bases as data sources. {% embed url="" %} To see the Knowledge Articles that have been ingested into your Tenant, navigate to SOR > Knowledge. # Ontology {% embed url="" %} ### Add Ontology {% embed url="" %} # Prompts Studio {% embed url="" %} ### Create a Prompt {% embed url="" %} After you create a prompt you see the Prompt Playground, where you can continue to edit the prompts. ### Prompt Playground

# Requests Window You can monitor requests sent to your bot and review requests individually. You can set up permissions for Aisera Admin UI Users to perform different tasks, such as **Review**. {% embed url="" %} # Similar Documents {% embed url="" %} # Tickets To see the **Tickets** that have been ingested into your Tenant, navigate to **SOR > Tickets**. {% embed url="" %} # Workflows - Tenant ### Browse Workflows Library {% embed url="" %} ### Add New Workflow {% embed url="" %} # Prerequisites Aisera sells Gen AI/Agentic applications/bots. You can purchase these applications/bots in different forms, known as Aisera products.

The products are sold in 2 different deployment styles:\ 1\. **Aisera Team Deployment:** where your Aisera team builds the application/bot end-to-end. 2. **Customer Deployment:** where you use the Aisera Administration Application (also called the Aisera Admin UI) to create most of the application and deployment yourself, with some assistance from your Aisera Onboarding Team.\ \ If you are performing your own application/bot build and deployments, Aisera will give you a tenant URL that will open your company's instance of the Aisera Admin UI.

## Enterprise Applications Aisera products are deployed within Enterprise applications that your company already owns.

## Customer Users You can embed a chatbot within your customer support portal so that it can answer questions based on information in your Support Knowledge Base.

The screenshot above shows the [Aisera Support Portal](https://support.aisera.com) with a chatbot that our customer can access if they have sign-in credentials for the Support Portal. Using the chatbot deflects ticket creation by allowing customers to get their questions answered instantly without the need to open a ticket. ## Employee Users Aisera also uses a chatbot, that we deploy internally for our Customer Support Engineers as part of the Agent Assist product. With the Agent Assist product, human agents can see similar, previous tickets while they're working on new tickets. At the same time, they can use a chatbot assistant that has information from 3 different Enterprise applications, used internally at Aisera that provide the information for the Support Knowledge Base.

The screenshot above shows an example implementation of Agent Assist where the Agent Assist application has been added to a ServiceNow Enterprise application (internal ServiceNow application used by employees within a specific company). ## Access to Your Enterprise Application Administrator In the first instance above, Aisera owns the Zendesk Enterprise application, which is where we have integrated the chatbot that we use on our Support Portal. In the second example, a customer has integrated Agent Assist into their ServiceNow Enterprise application. Whether you plan to deploy your Aisera application/bot for use by your internal employees or add it to an external facing portal for use by your customers, it is essential that you allow the people deploying your bot to interact with the Administrator for your Enterprise software.

This may be an AWS or other cloud administrator if your software is administered off-site, or your IT department if it is administered locally. ## Putting the Bot to Work Aside from answering questions based on information in your Tickets and Knowledge Bases, your Aisera applications/bots can create tickets, sort tickets, track incidents, and generate new knowledge articles based on existing information. The Aisera Admin UI allows you to create Events, edit Prompts, and design Workflows to accomplish advanced tasks. # Setup Your Aisera Tenant Each Aisera customer has an instance of the Aisera Gen AI Platform, called a **Tenant**. Each tenant runs in the Aisera Cloud, and each tenant has a separate URL address. This tenant will be accessed through the **Vanity** **URLs** provided to you by Aisera.

The Aisera Gen AI Platform uses a **multi-tenant architecture**, which means that for each customer, there is a separate container with server software, a temporary data store , and some shared processing modules. Your Aisera Team will assign you a **Tenant ID** that will define your tenant instance. Use the **Tenant ID** when you log into your Aisera Gen AI Platform instance.

## Multiple Bots Per Tenant Instance Many companies create a single bot for a single purpose that uses the resources of their tenant. But you can create multiple bots for your company tenant instance. For instance, you may want an IT bot to help with IT requests, and an HR bot to help with HR requests.

Bots will share tenant parameters, but each bot has its own **Bot ID** and unique parameters. ## Universal Bot If your company wants to use more than one bot, you can create separate bots or create a **Universal Bot** that connects them.

## Adding Data to your Tenant To connect a Data Source with your Aisera Tenant Instance, you need to perform these connector steps: 1. Create an **Integration** with your Aisera Tenant (provide authorization) 2. Include your **Integration** in a **Data Source c**onfiguration (data from your Enterprise application) 3. Associate the **Data Source** with an **Aisera Application/Bot**. This is one of the steps you will perform when you get to the Build-a-Bot stage. ### Integrate a Data Source with Your Tenant The Aisera term **Integration** means to create a way for either an Aisera Admin user to log onto your data source and bring data to the temporary data store on your Aisera tenant, or for an Admin user of your data source to log onto your Aisera tenant and transfer data to the temporary data store on your tenant. Setting up the authentication credentials for this exchange is known as **Creating the Integration**. To begin, navigate to **Settings > Integrations** in your Aisera Admin UI and you will see that many commercial sources are pre-mapped to the Aisera Platform.

Choose the **+ New Integration** button in the upper-right corner of the **Integrations** window. \ The **New Integration Search** window allows you to choose our list of pre-mapped external data sources by category, like **Cloud Providers** or by Enterprise application name.

Step-by-step instructions to connect many of these sources is documented [here](/aisera-platform/adding-data-to-your-tenant/integrations-and-data-sources/connectors). If the source you're connecting to is not documented yet, find out the authentication type you need to use, and follow [these general instructions](/aisera-platform/adding-data-to-your-tenant/integrations-and-data-sources/learning-functions-for-external-data-sources/generic). ### Add the Integration to a New Data Source Now you need to include this new Integration in a New Data source (Enterprise application) that you add to your Tenant configuration. Navigate to **Settings > Data Sources** in your Aisera Admin UI.

Choose the **+ New Data Source** button in the upper-right corner of the **Data Sources** window. You will see the same list of pre-mapped external data sources. The **New Data Source Search** window allows you to choose from the same pre-mapped external data sources by category, like **Cloud Providers** or by Enterprise application name. Make sure that you choose the same source for your Data that you chose for your Integration (they must match, because they are two pieces of the same connection). For instance, if the Enterprise application that your company uses to track Tickets and Knowledge Articles is Salesforce, *chose Salesforce for your new **Integration**, and your **Data Source**.* If the source you want to add is not in the **Settings > Data Source > + New Data Source** list, you can use a [**Generic** **Integration**](/aisera-platform/adding-data-to-your-tenant/integrations-and-data-sources/using-the-generic-connector) and then use [**Field Mapping**](/aisera-platform/adding-data-to-your-tenant/data-ingestion#field-mapping). #### Map Enterprise Data to your Aisera Platform The term Data Source for the Aisera Gen AI Platform is usually the Enterprise application that your company uses to create shared information, such as Tickets or Knowledge Base Articles that the Aisera platform will ingest and keep in temporary storage.

A few of the common Enterprise software sources that our customers use are Salesforce, ServiceNow, Zendesk, and Confluence. #### Mapped Fields The term Mapped Fields in the Aisera Gen AI Platform means that there is a software map where the Aisera Gen AI Platform, your Aisera Team, or you (with the Aisera Admin UI) have created a one-to-one correspondence between the fields in the Aisera Gen AI Platform's **Data Model** and the fields in your Ticket or Knowledge Tools. For instance, if you plan to use Tickets from your Salesforce tool as a source of information for your bot responses, then you need to create an Integration for your Salesforce instance so you can ingest the data from the Tickets table and the other tables in Salesforce that are related to the Tickets table (to make a complete Ticket).\ \ Then you will Add a Salesforce Data Source to your tenant, and this will connect the pre-mapped Salesforce fields with the fields in the Aisera Gen AI Platform so that the data can be transferred to the temporary data store on your tenant instance.\ \ The example below displays the names of Aisera Knowledge Base Article fields on the left side and the corresponding Salesforce fields for Knowledge Base Articles on the right.

You can see the table above on the Data Source Details page of the Aisera Admin UI after you run a Data Ingestion Job, as described below. #### Ingest the Data After you have mapped your Data Source with the Aisera Gen AI Platform fields (either using the Aisera Admin UI or with a JSON file), the data will automatically be ingested when your Data Source configuration is complete. If you need to run a Data Ingestion Job manually, use the triangle button at the top of the Data Sources window to copy the data from your source to your Aisera tenant.\ \ This completes the Connection for your Tenant! ### Associate the Tenant Data Source with a Bot The last step in the data connection process is associating the Data Source with your Bot. You can do this any time AFTER you run a Data Ingestion Job, but BEFORE you complete your Bot configuration. After you have performed the initial steps of bot creation, choose the

button to associate the tenant data source with your bot. Choosing the tenant data source for your bot allows the bot to use the data from the connected source. #### Complete the Tenant Setup by Adding Channels Adding channels to your tenant is that last thing you have to do before you start to build your application/bot. You can perform this step at the beginning, or come back and do it later. ## Tenant Configuration Options There are many tenant configuration options in the Aisera Gen AI Platform. If you suspect that something in your tenant configuration is not set correctly, you can check the [**Tenant Configuration Options**](/aisera-platform/tenant-setup/aisera-platform-configuration/tenant-configuration-settings). # Bot Essentials The Aisera Gen AI Platform is a cloud hosted software, meaning it has server software running in the cloud that interacts with your client application.

### Login Credentials You can set login credentials for your bot when you create it. ## Bot Channels The default bot application is a Webchat bot, but other applications are available that have specialized functions and features.

The 'look and feel' of your bot is known as a **Channel** in Aisera Gen AI terminology. \ \ The channel determines how your end users will interact with your Aisera Application, whether it is through a webchat as shown above or through a messaging application like Microsoft Teams.

The screenshot above shows an application/bot with a Copilot channel. You can set the configurations for each channel in the Channel Details window, as shown below.

In this example, the login screen appears on the Copilot bot because the Authentication Type for the bot is set to Email. ## Bot Conversations An AiseraGPT (bot or application) can engage in conversations as well as derive information from data sources to serve answers to customer requests. ### Enterprise Domain Packs The type of conversation you want your bot to have is determined by the **Domain** or category of information that you want your bot to use. For instance, in the screenshot of the default bot above, the bot is taking a food order for pizza. You might refer to this as the Restaurant, Hospitality, or Pizza Order domain. The type of application that you build with the Aisera Gen AI Platform depends on the **Enterprise Domain Pack (EDP)** that you choose when creating your application and the **Channel** or 'skin' that your application uses. When you create an **AiseraGPT** Bot, the bot creation wizard will ask you what Domain you want to use for your bot.

For instance, if you want to create an **Aisera Assist** bot for your IT department, you would choose the **IT** value for the **Enterprise Domain Pack** and then create a description or purpose for your bot. The **EDP** that you choose for your application or bot determines the default **Conversations** and **Intents** that it will use. ### Conversations and Intents The EDP that you choose for your application or bot determines the default **Conversations** and **Intents** that it will use. Choose the **View Conversations** button on your application's Details page to see the conversations that are included with the EDP that you have chosen.

Conversations Included by Default for a Specific EDP

Select the **View Intents** button on your application's **Details** page to see the **Intents** that are included by default with the EDP that you have chosen.

When you select a specific **Intent**, you can see more information and make customizations in the **Intent Details** window.

### Requests, Intents, and Fulfillment in Workflows Each conversation with the chatbot takes in a Request, determines the Intent of the Request, and provides Fulfillment for that Request as a Response. In it's simplest form, there is a one-to-one correspondence between the question (request) and the answer (response).

In the world on Gen AI, you associate the Request with an Intent so that the chatbot can fulfill the request in the best way possible, even if there are different ways to make the same request. For instance, the requests 'Do you know the time?' and 'What time is it?' have the same Intent. But the Requests, 'Do you know what time my meeting is?' and 'What time is it?' have different Intents. As you can see from the diagram below, the Fulfillment can take different forms, depending on the Intent of the Request.

In the Aisera platform, you set the conversation and action sequences for your application or bot using Workflows that you build with the AI Workflow Studio and then associate with a specific application. A Hyperflow is an advanced type of workflow that allows you to build a flow based on the integrations that are connected with your application or bot. ## Intentless Requests with RAG Answers in Hyperflows The Aisera platform also supports Requests that are more complicated, and may include an Intentless goal that requires a set of Actions and RAG indexing of your documents to complete.

\

# Who Does What? At the beginning of your Aisera implementation, there are a few different Admin Personas involved in the setup - some at your site and some on your Aisera Team. ## Aisera Administrator The Aisera employee who gives you your tenant endpoint and sets up your tenant (server) instance, is an **Aisera Administrator**. This is the person who can send commands directly to your Aisera server and temporary data store. While you're onboarding, you may work with more than one **Aisera Administrator** to complete your setup. ## Enterprise Administrator The Administrator for the Enterprise software that you want to integrate with your bot is the most important person (and sometimes the hardest to find or access) in the bot builder journey.\ \ The Enterprise software administrator (for your company-owned applications, like ServiceNow, Salesforce, Zendesk, SharePoint, Confluence, Slack, MS Teams, or other internal tools is important because they need to give your Aisera platform the same access to your Enterprise application that a user at your company would have. For instance, when a user logs in to your Enterprise application, they use specific credentials, based on the type of authentication you use for your company software. The Aisera Gen AI platform needs to use the same authorization type to log into your Enterprise application. This is why your Enterprise application Admin needs to create authentication credentials for your Aisera Gen AI platform before the platform can integrate with your Enterprise software. This allows the Aisera application to log in and access **Ticket** and **Knowledge Base** data to fulfill the application/bot **Requests**.\ \ NOTE: If the authentication type you're using does not come up in the Integration wizard for the Enterprise application that you're using, you can choose from the **Supported Authentication Types** after choosing the **Generic Connector**. ## Customer Administrator The **Aisera Customer Administrator** is you.

The person who is creating, designing the workflow, training, and maintaining your Aisera Gen AI Bot. This Aisera Product Documentation was created to help you success with your tasks, using the Aisera Administration Application (Aisera Admin UI).\ \ Your Aisera Team will collaborate with you on the design and onboarding of your Aisera Application/Bot. One of the first tasks you'll take on together are the use cases for your bot - what do you want it to do, and what problems do your want it to solve? After the Aisera Administrator gives you a tenant endpoint, you can enter the endpoint address in a browser window, and you will see the Aisera Administration Application (also known as the Aisera Admin UI). ## Administrator Tasks The following diagram will give you an idea of the personas who usually perform specific tasks when you start to create your bot. Before you can build an Aisera Application/Bot, your Aisera Team will work with you to: * Set up your [**Tenant**](https://docs.aisera.com/overview-of-aisera/getting-started-guide/setup-your-aisera-tenant) and [**Tenant Configurations**](https://docs.aisera.com/aisera-platform/tenant-setup/aisera-platform-configuration/tenant-configuration-settings) (depends on Use Cases) * Create the Customer Administrator and other Aisera Admin UI [**User Roles**](https://docs.aisera.com/aisera-platform/tenant-setup/aisera-platform-configuration/aisera-platform-roles-and-permissions) * Upload Bot Users via File Transfer. After upload, they will appear on the Users window within the Aisera Admin UI. * Set Up [**Integrations**](https://docs.aisera.com/aisera-platform/adding-data-to-your-tenant/integrations-and-data-sources/creating-a-connector/adding-and-managing-integrations-in-your-tenant) and Map [**Data Sources**](https://docs.aisera.com/aisera-platform/adding-data-to-your-tenant/integrations-and-data-sources/creating-a-connector/adding-and-managing-data-sources-in-your-tenant) to Your Tenant * [**Ingest Data**](https://docs.aisera.com/aisera-platform/adding-data-to-your-tenant/data-ingestion) from your Data Sources to Your Tenant * Upload [**Ontologies**](https://docs.aisera.com/aisera-platform/llm-operations/ontology) Then you'll be set to create your bot(s). Now you'll start making choices and creative decisions based on your **Use Cases**. Your Aisera Team will work with you to: * Modify the Default Bot [**Conversation Phrases**](https://docs.aisera.com/aisera-platform/crafting-the-conversation/configuring-icm-intents-and-fulfillment#heading-h.3as4poj) and Prompts (set by Domain choice)\ (What Your Bot Says) * Use the [**Requests Window**](https://docs.aisera.com/aisera-platform/ai-workbench-optimize-conversations/requests-window) * Use the [**Intents Window**](https://docs.aisera.com/aisera-platform/crafting-the-conversation/configuring-icm-intents-and-fulfillment#all-intents) * Use the [**Prompts Studio**](https://docs.aisera.com/aisera-platform/llm-operations/prompts-studio) * Create Workflows and Hyperflows\ (What Your Bot Does) * Use the [**AI Workflow Studio**](https://docs.aisera.com/aisera-platform/ai-automation-optimize-flow/ai-workflow-studio) drag-and-drop canvas * Use [**Event Studio**](https://docs.aisera.com/aisera-platform/ai-automation-optimize-flow/event-studio) to create Event Triggers/Handlers * Send [**API Requests**](https://docs.aisera.com/apis) via Workflows * Use [**Test Suite**](https://docs.aisera.com/aisera-platform/ai-workbench-optimize-conversations/test-suite) Tools * Iteratively Test and Train Your Bot * Use [**AI Workbench**](https://docs.aisera.com/aisera-platform/ai-workbench-optimize-conversations) to analyze Request/Fulfillment Results View ## Application/Bot User The application/bot user is the person who interacts with the application/bots that you create with the Aisera Admin UI. It is important to remember the perspective of these users when you create conversation options and parameters for your applications/bots. The data for people who will use the application/bot must be uploaded to the Aisera GenAI platform via a data file. After the file us uploaded, you can view the user data in the **Users** tab near the top of the left navigation menu within the Aisera Admin UI.\ \ Do not confuse these users with the users you can view with the **Settings > User Accounts** command. Those are users of the Aisera Admin UI. # Using the Aisera Admin UI When you purchase an Aisera Gen AI Product, your company receives an endpoint (URL) that opens a specific instance of the Aisera Gen AI Platform in the cloud. This is called your Aisera tenant instance and it is your company's Aisera software environment.

Your Aisera Company Admin and your Aisera Onboarding Team will navigate to this URL and use the Aisera Administrator Interface (also known as the Aisera Admin UI) to create Agent Assistant Bots that you can embed in one of your corporate enterprise applications to help perform frequent requests and answer commonly asked questions. ### Logout Button Use the Logout button in the upper-right corner of the Aisera Admin UI to ensure that you completely close your browser session after using the Aisera Admin UI. ![](https://docs.aisera.com/~gitbook/image?url=https%3A%2F%2F2983236984-files.gitbook.io%2F%7E%2Ffiles%2Fv0%2Fb%2Fgitbook-x-prod.appspot.com%2Fo%2Fspaces%252FiZkLJr3EjXkd2tHYiQJP%252Fuploads%252Fo4aUmvmWaFljFPdeI15e%252F2.png%3Falt%3Dmedia\&width=768\&dpr=4\&quality=100\&sign=ba1a538\&sv=2) **To logout:** 1. Navigate to your user name in the upper-right corner of your Aisera Admin UI. 2. Select **Logout** from the pull-down menu. ## Use Cases Most of the applications/bots that our customers build ingest data from either **Tickets** (that are created when agents communicate with customers) or from **Knowledge Base Articles** or similar documents that contain useful information for customers. Therefore, common use cases for bots created with the Aisera Gen AI platform include actions performed on Ticket data or Knowledge Article data: **Agent Assist** - Comparing new ticket information in real time (while agents are working with customers) with existing similar tickets. **Ticket Concierge** - Sorting tickets into categories. **Generating Knowledge from Tickets** - Creating Knowledge Articles from ingested Ticket Data. **Generating Knowledge from Ticket Comments** - Creating Knowledge Articles from ingested Ticket Comment fields. **Knowledge Learning** - Using LLM models to provide answers based on ingested Knowledge Articles. **User Learning** - Collecting and reusing information about the application/bot users. ## Aisera Admin UI Command Reference The following sections describe the main windows in the Aisera Admin UI, and the commands you can perform within each window. The Aisera Admin UI contains two main menus - an outer menu and an inner menu. Starting with the outer menu, this reference shows you the available windows with links to information about usage of each window.

# Styling your Aisera Webchat Customizing the Webchat Channel to fit your Brand ## Overview The Aisera GenAI platform offers a customizable [Webchat](/aisera-platform/channels/aisera-webchat) channel that allows users to interact with bots. This default webchat includes various components that can be tailored using CSS to match your brand's style seamlessly. For more information about setting up a web channel, see also: [**Add a Web Channel**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/add-a-channel) and [**Customize Your AiseraGPT**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/customize-the-look-and-feel-of-your-default-webchat-bot). ## Webchat Components The following diagram breaks down the visual components of the Aisera Webchat interface. use thee numbered legend below to identify specific elements of the webchat.

1. Header 2. Channel Logo 3. Chatbot Name 4. More 5. Resize and Collapse Controls 6. Message Banner 7. Bot Icon 8. Initial Greeting Message 9. Initial Action Recommendations 10. Footer Aisera Logo 11. Opener Icon Image 12. Opener Message 13. Opener Message Placement ## Customizing the Webchat To customize the look of your webchat navigate to **Channels > Your Channel** to edit the appearance of your webchat channel. There are two modes or editing the webchat channel: **Standard:** This mode simplifies customization by offering a guided experience. You can effortlessly personalize their webchat by filling in standard fields for each component, ensuring a quick setup. **CSS Mode:** This mode enables you to edit the webchat's CSS directly, offering customization flexibility but possibly resulting in a less cohesive appearance. ## Standard Mode Fields ### Opener Icon This is the imaged displayed for the opener of the webchat. Ensure this image is at least 50x50 pixels for clear visibility. ### Opener Background Color This changes the background of the chat opener container.

### Opener Size This sets the size of the opener in pixels. The opener adjusts content to fit its size. Larger images will shrink, and smaller ones will enlarge, possibly causing blurriness. ### Opener Placement Determines which corner of the viewport the opener use as the origin. {% hint style="info" %} The opener is placed on your webpage relative to the viewport. It is best practice to ensure that the opener does not obscure content or interfere with other actions on your webpage. {% endhint %} ### Opener Right/Left This value sets how far the opener is offset from the edge of the viewport horizontally. ### Opener Bottom This value sets how far the opener is offset from the edge of the viewport vertically. ### Allow Removal This enables a button in the header of the webchat that allows users to hide both the webchat and the opener from the webpage. If this button is pressed, the webchat and opener will be hidden for 24 hours at which point the opener will automatically begin showing again.

If this configuration is enabled, the host webpage should then provide a way to show the opener or webchat at the user's request. This can be done using either the `aisera.webchat.show` message, or the `aisera.webchat.open message`. The following is an example of a button which would show the webchat again: ```html ``` ### Opener Message Time Sets the time before the **Opener Message** is displayed if the webchat has not been opened during that time. This must contain a value is there is an Opener Message. ### Opener Message The message displayed to a user after as set amount of time, nudging the user to interact with your bot. This message only appears if the webchat has not been opened before the set amount of time has passed, and will not display more than once if the page is not reloaded.

### Opener Message Closing Time This sets the amount of time to hide the **Opener Message** if the **Opener Message** is closed by a user. After the period of time has elapsed the opener message will begin displaying again. ### Hide Webchat Resize This configuration hides the webchat resize button.

### Minimized Window Size This sets the size of the Webchat window while the Webchat is minimized. The **Width** is set in pixels. The **Height** is set as a percentage of the total vertical space within the viewport minus the offset in [Opener Bottom](#opener-bottom). ### Maximized Window Size This sets the size of the Webchat window while the Webchat is minimized. The **Width** is set in pixels. The **Height** is set as a percentage of the total vertical space within the viewport minus the offset in [Opener Bottom](#opener-bottom). ### Chatbot Name This field changes the name of the bot displayed in the header of the Webchat.

### Logo This allows you to change the logo displayed next to the **Chatbot Name** in the header. This image will be scaled to to have a maximum height of 26 pixels while maintaining the aspect ratio of the original image. {% hint style="info" %} If you provide a logo image but no **Chatbot Name** only the logo will appear. This allows you to supply an image containing both your logo and brand name with your desired typography. {% endhint %} ### Logo Alt Text This field defined the alternative text for the brand logo image. This string is read by screen readers and is displayed if the image fails to load. {% hint style="info" %} Avoid using "Logo of..." or "Image of..." in the alt text. Screen readers will announce the element as a graphic. Simply use "\" for this text. {% endhint %} ### Header Background Color This field sets the hex code for the header background color. {% hint style="info" %} Ensure the header background has a contrast ratio of at least 4.5:1 against the [header text color](#header-text-color) for [accessibility compliance](https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html). {% endhint %} ### Header Text Color This fields sets the hex code for the text color in the header. This will also affect the color of the default Icons within the header. {% hint style="info" %} Ensure the text color has a contrast ratio of at least 4.5:1 against the [header background color](#header-background-color) for [accessibility compliance](https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html). {% endhint %} ### Border Color Sets the border color between the header, message list, and chat box areas.

### Icon Fields The icon fields allow you to change the default icons for each operation. This can be done by uploading an image file to be used in place of the default icon. Icons will be scaled to a fixed size so it is recommended that the image input is of a similar size to avoid images becoming blurry or losing find detail. * Maximize Icon * Minimize Icon * Collapse Icon * More Menu * Refresh Session Icon * Sound Off Icon * Sound On Icon * Email Transcript Icon * Logout Icon ### Link Color This field sets the hex code for all hyperlinks served to a user by the chatbot. {% hint style="info" %} Ensure the link color has a contrast ratio of at least 4.5:1 against the chat bubble background for [accessibility compliance](https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html). {% endhint %} ### Required Asterisk Determines the visibility of the asterisk, used as a "required" indicator, on input labels when filling out forms within the chat bot. ### Bot Icon This field sets the image displayed next to your bot's messages in the chat message list. This image has a fixed width of **36px**. Uploaded images should have a **1:1** aspect ratio to avoid distortion. To prevent detail loss when scaling down small images, it's advisable to use a simplified logo or mascot. ### User Background Color This field sets the hex code for the background of the user chat messages in the message list. {% hint style="info" %} Ensure the user background color has a contrast ratio of at least 4.5:1 against the [user text color](#user-text-color) for [accessibility compliance](https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html). {% endhint %} ### User Text Color This field sets the hex color code for the text contained in the user chat messages in the message list. {% hint style="info" %} Ensure the user background color has a contrast ratio of at least 4.5:1 against the [user background color](#user-background-color) for [accessibility compliance](https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html). {% endhint %} ### Live Agent Icon This field sets avatar image displayed when a live agent joins the conversation or takes over for the bot. This image has a fixed width of **36px**. Uploaded images should have a **1:1** aspect ratio to avoid distortion. To prevent detail loss when scaling down small images, it's advisable to use a simplified graphics. ### Show Input Filed This field determines the *visibility* of the text input allowing users to enter arbitrary requests. {% hint style="warning" %} This affects the HTML element's *visibility* only and does not prevent arbitrary requests from being submitted to the webchat. {% endhint %} ### Show Aisera Logo This field determines the visibility of the "Powered by Aisera" logo in the footer of the chat window. ### Footer Message This field allows you to add a supplementary message to the footer. This field allows a maximum of 200 characters and supports bold, italic, and hyperlink formatting. ### Send Icon This field allows you to replace the default paper airplane send icon used to submit text to the bot. The image dimensions will be set to **20x20px**. ## Best Practices Here are best practices for styling your webchat. Adhering to these will enhance the appearance and functionality of your webchat. ### Use of Color Hues To maintain a professional and cohesive interface, limit your color palette to three to four color hues. Over-complicating the color scheme can become distracting an decrease readability within the webchat. ### Text Contrast and Accessibility To ensure your chat interface is accessible for everyone, including individuals with visual impairments or those in high-glare environments, maintain a contrast ratio of **4.5:1** between text and the background of the text container. Learn more about this from the [Web Content Accessibility Guidelines (WCAG)](https://www.w3.org/WAI/WCAG22/Understanding/contrast-minimum.html). ### Font Size and Legibility Maintain a minimum body font size of **15px** to ensure the conversation remains readable for all users across mobile and desktop devices. ### Opener Icon Size For optimal clarity and visibility, ensure that the image designated for the opener icon is no smaller than 50x50 pixels in size. This will enhance the visual quality and ensure that the icon remains distinct and easily recognizable to users. # Tour of the Documentation The video below gives you a brief tour of the Product Documentation.

[**Click for Video**](https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FMSMLfalqmdzhbdB7V15d%2Fuploads%2FPbx41JC4QiZK5CJAN7rw%2Ftour_of_the_docs.mp4?alt=media\&token=63612247-d10b-4ce6-ae30-78816872b9c1) Note that this video is temporary and will be updated in a future release. # Overview of the AiseraGPT App (Bot) [AiseraGPT](https://aisera.com/products/aiseragpt/) is Aisera's generative AI solution tailored for enterprise environments, designed to elevate user experiences through intelligent automation and advanced natural language understanding. The term **AiseraGPT** is used within the Aisera Admin UI to represent a generic application/bot (similar to ChatGPT) that you create and build upon to make an AI Agent that is specific for your Enterprise configuration and use case. You configure your agent by connecting it with your **Data Source** and then customize it using **Channels**. At its core, AiseraGPT leverages [domain-specific large language models](https://aisera.com/blog/domain-specific-llm/) (LLMs) that are grounded in an organization's unique data. This ensures that the responses generated are not only accurate but also contextually relevant, addressing the specific needs and nuances of different business sectors such as IT, HR, Finance, and Customer Service. #### Key Features **1. Contextual Disambiguation:** AiseraGPT excels in contextual disambiguation, which is critical for understanding and correctly interpreting user queries that may have multiple meanings. This capability allows the system to consider the context in which a query is made, including the user's past interactions, current activities, and specific terminologies relevant to the organization. By doing so, AiseraGPT can deliver precise and contextually appropriate responses, significantly reducing misunderstandings and enhancing the overall user experience. In case the system does not have enough information to provide a response, if the user asks "My computer is broken" for example, it will consider the contextual history at its disposal to ask disambiguating/clarifying question until the user's intent is clear. **2. Multi-Intent Handling:** One of the standout features of AiseraGPT is its ability to handle multi-intent queries effectively. In many cases, users may have multiple requests or questions embedded in a single interaction. Traditional systems struggle with such complexity, often addressing only one aspect of the query. AiseraGPT, however, can parse and understand multiple intents within a single query, providing comprehensive responses that address all aspects of the user's request simultaneously. This multi-intent handling capability streamlines interactions, making them more efficient and user-friendly. **3. Multi Fulfillment:** Under certain circumstances, AiseraGPT is able to provide the user with multiple fulfillment options e.g. a combination of Workflow, Knowledge Article and Service Catalog. In general, it identifies the user's intent and provides knowledge based answer if the user is looking for information or workflow driven action if the user is making a request. However, the system will provide the adjacent fulfillment if that is important for the user. For example, if the user asks how do I change my password, AiseraGPT will respond with a knowledge based answer while also providing the user with the ability to change their password through a workflow **4. Unified Interface with Universal Bot:** AiseraGPT includes a Universal Bot that offers a unified interface for handling user requests. Unlike conventional single-purpose, rule-based bots, this AI-driven bot provides ChatGPT-like responses across multiple domains, channels, and languages. The Universal Bot intelligently routes requests to the most appropriate bot or domain specific application, ensuring that users receive the best possible resolution quickly and efficiently. The Universal Bot supports seamless communication across various channels, enhancing accessibility and convenience for users. **5. Real-Time Summarization and Answer Generation:** AiseraGPT is equipped with advanced capabilities for generating real-time summaries and contextual answers. By extracting and analyzing information from all available data sources and business applications, it can provide detailed yet concise responses tailored to user preferences and past interactions. This feature not only saves time for users but also ensures that they have access to the most relevant information without having to sift through large amounts of data. The system is also able to leverage white listed public sources of knowledge (e.g. microsoft.com/support) to provide answers to user questions as well as commingle the public information with private organization specific knowledge. **6. AI Workflows and Automation:** The platform’s ability to trigger [AI workflows](https://aisera.com/blog/ai-workflow-orchestration-guide/) automates a wide range of tasks, from simple queries to complex processes. This automation significantly boosts productivity by reducing the need for manual intervention and allowing users to focus on more strategic activities. Whether it’s resetting passwords, updating user permissions, or generating detailed reports, AiseraGPT handles these tasks with precision and speed, taking action against multiple enterprise applications and services. **7. Integration with Existing Systems:** With more than 500 out of the box connectors, AiseraGPT seamlessly integrates with a multitude of enterprise applications including ITSM systems. This extensive integration capability ensures that AiseraGPT can be easily incorporated into an organization's existing IT infrastructure, enhancing its functionality without requiring significant changes or disruptions. **8. Personalized User Experience:** Through continuous learning and adaptation, AiseraGPT offers a highly personalized user experience. It takes into account individual user behavior, preferences, and historical data to tailor interactions. This personalization leads to higher satisfaction rates as users feel that their unique needs and contexts are understood and addressed effectively. **9. Enhanced Security and Compliance:** Built with enterprise-grade security in mind, AiseraGPT adheres to stringent security protocols and compliance standards. This ensures that all data interactions are secure, protecting sensitive information and maintaining user trust. The AI Experience (AIX) platform on which AiseraGPT is built incorporates features like encrypted data storage, secure access controls, and comprehensive audit logs. #### Impact on Business Operations By implementing AiseraGPT, organizations can achieve significant operational efficiencies and cost savings. The AI-driven automation and advanced analytics capabilities lead to faster issue resolution, reduced operational costs, and improved productivity. For example, the system's ability to resolve common IT and customer service inquiries autonomously allows human agents to focus on more complex and value-added tasks, ultimately enhancing the overall service quality and user satisfaction. AiseraGPT's transformative impact extends across various industries, demonstrating its versatility and effectiveness in addressing diverse business needs. Whether in financial services, healthcare, retail, or technology, AiseraGPT provides a robust AI solution that empowers organizations to deliver superior service experiences, drive innovation, and maintain a competitive edge in their respective markets. For more detailed information, please visit [Aisera's official page on AiseraGPT](https://aisera.com/products/aiseragpt/). # How to set up an AiseraGPT Bot The topics in this section describe how to create an Aisera virtual assistant/bot.

You will be able to create and see your new virtual assistant, even if you haven't established connections or integrated your systems yet. * [**Configure an AiseraGPT**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/configure-an-aiseragpt-bot) * [**Add a Data Source**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/add-a-data-source) * [**View Policy**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/view-policy-1.0-only) * [**Add a Channel**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/add-a-channel) * [**Add a Live Agent**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/add-a-live-agent) * [**Add an Error Handling Flow**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/add-an-error-handling-flow) * [**View App Conversation Messages**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/view-conversation-messages) * [**View App Intents**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/view-intents) * [**Assign Access Attributes**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/add-application-access-attributes) * [**Test Your Completed App**](/aiseragpt/how-to-set-up-an-aiseragpt-bot/test-your-completed-aiseragpt) # Configure an AiseraGPT Bot The term **AiseraGPT** stands for Aisera Generative Pre-Trained Transformer. It is used as a general term to refer to an Aisera **application** or **bot**. \ \ First you create a 'generic' application, and then you add the specifics by adding Channels and data Sources to your bot. ## Creating a Bot Use the following steps to create a virtual assistant. You will be able to create and see your new virtual assistant, even if you haven't established connections or integrated your data sources yet. To create a bot: 1. Login to the Aisera platform. 2. Select **Settings > AiseraGPT**

3. On the AiseraGPT page, click on + **New AiseraGPT** button on the top right hand side to launch the app configuration modal.

4. Fill in the required fields and use the default values for the remaining fields. Use the table below as a reference while you fill in the General Settings dialog Window.

Field	Description
AiseraGPT	Unique name of the application that will be displayed in the AISX interface. For example <YourCompany>GPT
Type	Determines if the bot will be Domain Specific or Universal Bot
Enterprise Domain Pack (EDP)	Each of the choices in the Enterprise Domain Pack (EDP) pull-down corresponds to a set of default Conversation Phrases and Intents that are already in the Aisera platform. You can add your own phrases and intents as you build out your assistant. Select the domain the bot will be operating in. This can be IT, HR, and so on.

Field

Description

AiseraGPT

Unique name of the application that will be displayed in the AISX interface. For example <YourCompany>GPT

Type

Determines if the bot will be Domain Specific or Universal Bot

Enterprise Domain Pack (EDP)

Each of the choices in the Enterprise Domain Pack (EDP) pull-down corresponds to a set of default Conversation Phrases and Intents that are already in the Aisera platform. You can add your own phrases and intents as you build out your assistant.

Select the domain the bot will be operating in. This can be IT, HR, and so on.

5. If you want to customize one of these domain packs by fine-tuning the LLM prompts before you add it to your virtual assistant, see the [**Prompts Studio**](/aisera-platform/llm-operations/prompts-studio). 6. Click the **OK** button to create the bot. Skip the following section and proceed to the **Summary/Details** section.

The following video shows you how to begin: [**Create a New Bot**](https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FksUDWGr9UTfOpa53PosP%2Fuploads%2FR0ufn11BI7dgTGkIOuPj%2Fcreate_a_new_bot.mp4?alt=media\&token=81546fca-34cf-42aa-8752-2121fd08594b)
### Advanced Settings Tab For this first bot, keep the default values in the **Advanced Settings** field. {% hint style="warning" %} You must fill in the **AiseraGPT, Type,** and **Enterprise Domain Pack (EDP)** fields before you can continue to the Advanced Tab. {% endhint %} #### Conversational AI 2.0 The Advanced settings window for your application/bot is set to Conversational AI 2.0 by default. This mean that your new bot can use the faster, intentless workflow and LLM capabilities.

{% hint style="info" %} Domain allowlist values (set by your Aisera admin) should be derived from your known applications. {% endhint %} #### Conversational AI 1.0 If you choose to turn off the Conversational AI 2.0 toggle in the Advanced tab, the bot you create will use the Conversational AI 1.0 features.

{% hint style="info" %} Domain allowlist values (set by your Aisera admin) should be derived from your known applications. {% endhint %} ## Summary/Details Window You will now see a **Details** (summary) window with your new bot settings, as shown below.

Note that the channel is already included, because the AiseraGPT bot uses the Aisera Webchat as the default channel.

### Adding a Data Source To allow the Bot to learn and utilize data for fulfillment, you must add a Data Source to the Aisera application. By default, Aisera applications do not have access to external data. Refer to the [connectors documentation](/aisera-platform/adding-data-to-your-tenant/integrations-and-data-sources) for guidance on linking data sources to the Aisera Platform. For details on how to ingest data, consult the [data ingestion documentation](/aisera-platform/adding-data-to-your-tenant/data-ingestion). To add a Data Source to your application: 1. Click the **+ Add Data Source** button to open the **Add Data Source** window. 2. Select the Data Sources you would like to add to your application. 3. Select **Ok** to add the Data Sources to your application. When a Data Source is added to your Application, the data ingested by that data source will become available for use by that application. #### Toggling Links Configure whether links from specific data sources appear in application responses. This control is Data Source specific, enabling you to manage link origins. The toggle for this is found in the Data Sources table under **HREF Links**. When this is enabled, links from this Data Source will be displayed in responses served to end users. Add a different channel if you want your bot content to show up within a different application, such as Slack or an enterprise application. Add the API channel if you want to create your own UI or add your bot to an existing UI using the Aisera APIs.

# Create a Universal Bot You can create a [Universal Bot](https://aisera.com/platform/universal-bot/) the same way you create any Aisera bot. 1. Navigate to **Settings > AiseraGPT** and then choose **+ New AiseraGPT**. 2. Create a **Name** for your application/bot. 3. Choose **Universal Bot** in the **Type** field.

4. In the **Advanced** tab, choose the **Domains** that your Universal bot will include.

5. Create a child bot for each of the **Domains** that are included in your Universal Bot.

6. Open your Universal bot and use the **+ Add AiseraGPT** button to add your child bots to your Universal bot. Select your child bots from the search list.

7. The child bots now appear on the **Details** page of your Universal bot.

# Universal Bot Analytics If you choose [Universal Bot](https://aisera.com/platform/universal-bot/) when you create an application/bot in the Aisera Gen AI Platform, you can create sub-bots that have different domains to perform specific agentic tasks. You can select your Universal bot from the bot list for your tenant (**Settings > AiseraGPT**) and then select **Analytics > Pre-Built Analytics** to see three consolidated views of insights for the sub-bots associated with your Universal app.
![](https://lh7-rt.googleusercontent.com/docsz/AD_4nXedCS8WwcKMR8HEdHjb0y4PtQMK3XUBdL6ez7WwngOJEA9FWxqEuUM6WlN_7x8gNYxK8P0J_0xatsKEkIkQfrb1nKPLCEWoVWM9VNpNqmUkhip15DD75m689KeQRryFsc_sgoXS9UiqDMDaScZv9Q?key=qlcwZD7a9HlKKbOKhWT_lQ) The Pre-built Analytics window for the Universal app consolidates all the KPI data from the different apps related to your Universal bot, on the Requests, Sessions, and Users tabs. Use the pull-down list to choose the bots that you want to include in the metrics data displayed on the Requests, Sessions, and User tabs.
![](https://lh7-rt.googleusercontent.com/docsz/AD_4nXejt3qGG3tpC3-ppUVowUbyq0dupmTA6JowGg0ojX4FSDuaQzdMoqvokVTMq8-mkYeLO7t1Ar6V_pZ81aDjNtEcMTbphUtwOHd7_IQmzyYEVXBpNas5Il_toHvbwW1wzW-GzmlAs_bafgZ3ZWq-RPk?key=qlcwZD7a9HlKKbOKhWT_lQ) \ You can choose to see the metrics only for your Universal bot, only for a child bot, or analytics data for all of the bots. # Add a Data Source ## Overview Now that you've created a virtual assistant, you need to have a source of information for your bot. The data source you use will depend on the purpose of your bot. When you set the EDP field for your bot, you chose the purpose/type of assistant that you wanted to create. ### To Add a Data Source: 1. While you are still viewing AiseraGPT Details Window, choose the **+ Add Data Source** button

2. After you click the **+ Add Data Source** button, you see the list of available data sources

3. Select the desired Data Sources and click **OK** 4. After you've added a Data Source, it will show up on the AiseraGPT **Details** Window for your bot. ### Make sure your EDP and Data Source are Compatible When you set the EDP field for your bot, you chose the purpose/type of assistant that you wanted to create. Make sure you choose a data source that corresponds with the EDP that you chose. | EDP | Possible Data Source | | ---------------- | ------------------------- | | IT | System or Process Manuals | | HR | Employee Handbook | | Customer Service | Knowledge Base Articles |

The following video shows you how to begin: [**Add a Data Source**](https://files.gitbook.com/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FksUDWGr9UTfOpa53PosP%2Fuploads%2FXQ6jGsEjJ3PjXAtlulVi%2Fadd_data_source_to_bot.mp4?alt=media\&token=28e2d055-97e6-44e5-bf0e-cde9858db411)
# View Conversational AI Policy If your bot uses the Conversational AI 2.0 option, the **View Policy** button is now called the View Conversational AI Policy button. The parameter on a 1.0 bot that was called, **Agent Persona Snippet** has been renamed to **Agent Response Format Instructions** and allows you to specify instructions for the 2.0 bot response.

The Aisera **Conversational AI** **2.0** bots support an app policy variable where custom instructions can be entered to fine tune the generated LLM output. These instructions can be appended to the standard generation prompt. This way, the change will not impact other bot behavior: * It will not be lost if the app is switched to a different LLM * It will not inadvertently impact other features or data * Any product updates to the prompt will be applied correctly to the app (regardless of the customization) ### Agent Response Format This variable allows you to provide context and tone for the conversation.\ \ **Example App Policy** - Please keep the tone formal. Never use bold or italics formatting. If the context includes any instructions, provide your response in a step-by-step manner using bullet points.

### Show Images in KB Output If your **Knowledge Base Articles** have images embedded in the content, then those images may be used in generated RAG responses that are sent to users on Webchat, Teams, or Slack channels.

If you do not wish to use images in generated answers, you can turn **OFF** this feature by unchecking the option. You might choose not to use images in your fulfillment answers if you find that images are not relevant to your application/bot users or if the image size changes the formatting in unexpected ways. In addition to the **Agent Response Format** and **Show Images in KB Output,** you can set your application/bot to work with ontologies, detect user countries, address users with profile information, use strict validation, and show images from KB Articles when relevant. # View Conversation Messages ### To View or Add Conversation Messages 1. Choose **Settings > AiseraGPT**, and then choose your bot from the virtual assistants shown on the page. 2. Click **View App Conversation Messages**

3. You now see the default conversation messages that your bot can use. If you’d like to use your own conversational phrases, choose the **Import** drop-down option from the **Actions** button, in the upper-right corner.

### To Import Conversation Messages The **Import** command will allow you to choose a file in JSON format to import. Select the **Paper Clip** icon to browse for the file you want to upload.

### To Export Conversation Messages If you’d like to export the conversation messages, choose the Export drop-down option from the **Actions** button in the upper-right corner. Select the checkboxes next to the phrases that you want to export, before you click the **OK** button.

### To Edit Conversation Messages To edit a conversation message click the **Pencil** icon on the right side of the entry to open the **Edit Conversation** modal.

You will see a dialog box that allows you to modify the message.

### Initial Conversation Messages The **View Conversation Messages** link on your application/bot **Details** page will take you to the **Conversation Messages** page. #### To set initial messages: 1. Type **`init`** into the **Search** field at the top of the page.

2. Review the list of **initial messages** and choose (or edit and then choose) the message that you want bot users to see on their first interaction with your application/bot. If you choose more than one initial message, the messages will alternate when new users interact with the bot. 3. Choose (or edit and then choose) an **initial recommendation** message for the bot users (optional).

4. Type **`toxic`** into the Search field at the top of the page. 5. Select (or edit and then select) the message that you want to appear if a user types an inappropriate message into your application/bot. ### **Enhanced Small Talk and Toxicity Option** Choose the **Handle Small Talk and Toxicity** option when you’re creating an **Aisera 2.0 Application/Bot** to override the previous settings with the updated Small Talk and Toxicity option.

This updated option incorporates the context of the conversation as well as the different search features available for each bot.
# Customize SSO Failure Message You can customize the message that users will see if they have a failure during an SSO login to your bot. To customize the SSO Failure Message: 1. Navigate to **Settings > AiseraGPT** and choose your bot. 2. In the **Channels** section of the Application **Details** screen, double-click on the Webchat Channel where you want to modify the message. 3. Choose the **Authorization** tab in the **Edit Channel** window. 4. Select the **SSO-SAML** radio button. 5. Add your custom message.

6. Click the **OK** button. # Unified Language Support As of the 5/26/2026 release, there are new parameters that determine the language in the conversation of the Webchat and AI Chat applications/bots. Similar to the Universal Bot, these channels now have unified language support. Currently, when you open an application/bot, Webchat or AI Chat, the bot starts the conversation in English, or the language you set as the application/bot’s default language, regardless of the language setting of the user’s browser. For instance, if a your browser is set to Spanish, the bot will still start the conversation in English. You can set the application/bot to match the language set in the bot user’s browser. #### **To set Auto Detection:** 1. Navigate to **Settings > AiseraGPT**. 2. Choose the application/bot that you want to modify. 3. Select the **pencil icon** to edit the application/bot. 4. On the **Details** page, scroll down the page to the **View Conversations** link. 5. Select the link. 6. Modify the [**initial conversation messages**](https://docs.aisera.com/aiseragpt/how-to-set-up-an-aiseragpt-bot/view-conversation-messages#initial-conversation-messages) to create or choose some initial messages in the languages your bot will support.

7. Go back to the application/bot **Details** page and scroll to the **Channels** section. 8. Select the **Webchat** or **AI Chat** channel.

9. On the **Edit Channel** window, select the **Use Browser Language** checkbox. **Live Language Assignment** If the user has not logged into the application/bot before, their user profile in the Aisera Gen AI platform is created to with the default language they have specified in their browser. **Localization of Error Messages** Error messages and other UI content will appear in the same language as the Webchat or AI Chat conversation. # Add Feedback Ability You can enable feedback to your bot. This will allow users to give feedback on how useful the responses are and if their issue was resolved. ## Enable Feedback You can add enable feedback by clicking the **Pencil** icon in the bot **Details** window and then selecting the **Not Helpful Feedback** checkbox in the General tab of the bot **Configuration** wizard.

Users will be able to add context when they select any of the following options as the reason for the **Thumbs Down** response: * Completely Inaccurate * Missing Relevant Information * Difficult to Follow * Something Else ## View Feedback You can view the analytics for the feedback and review items in the **Requests** tab of the **Pre-Built** **Analytics** window. To view feedback navigate to **Analytics > Pre-Built** in the left navigation bar.

### Unresolved Requests by Type Unresolved feedback can be viewed by type. This appears under the **Unresolved Requests by Type** on the **Requests** tab of the **Pre-Built Analytics** window.

### **Feedback Comment Column** The **Feedback Comment** column allows you to see the feedback comments for unresolved requests where the user has entered a **Thumbs Down** response, without clicking on each request to see feedback on the **Request Details** page. To see the Feedback Comment column: 1. Navigate to the **Analytics > Pre-Built Analytics > Requests** tab. 2. Click on **Not Helpful Rate** to get to the **Unresolved Requests** view. 3. Filter by **Resolution Type** and select **Not Helpful - Something Else**. 4. In the filtered view, click the E**llipsis** icon to **Add/Remove Columns**. 5. Add the **Feedback Comment** column.

See also: [**Conversation Resolution Analytics**](https://aisera.gitbook.io/aisera-product-documentation/05T7Y6PhrzglAXbQzVA8/aisera-platform/analytics/pre-built-analytics#requests-tab-conversation-resolution-analytics). ## Improving Fulfillment You can optimize and train your bot by modifying the fulfillment choices available to your bot to address issues that are unsolved. Choose a request in the list that has a **Resolution** value of **Unsolved**. The **Request Details** window has the option to show or hide **Not Helpful** options for the **Resolution Type**.

**Not Helpful** options are configured at the bot level, allowing you to selectively apply this option to specific bots. # Turn On Sound This sound option provides the ability for an Administrator to set the default parameter for sound notifications. Double-click the **Webchat** **Channel** on your AiseraGPT **Details** page. Under the settings for the Webchat channel, choose the **General: Introduce Notification Sound (Default)** setting with the following options: * Sound On * Mute All * Match Device Setting You can adjust this default setting according to your preferences.

[
](https://app.gitbook.com/o/c99bGZu7Q50seWWLiXHE/s/MSMLfalqmdzhbdB7V15d/application-messaging) # Connection Status All Aisera applications/bots contain a visual indicator when the chat connection is lost and reconnecting. The reconnect message gives you an indication of why messages are not being sent/received and lets you know when the connection is restored.

If the reconnection fails, the failure message is displayed. # Customize the look and feel of your Default Webchat Bot ## Viewing the Channel The Aisera Webchat channel can be tested and viewed from within the Aisera Admin UI. 1. Choose **Settings > AiseraGPT** and then choose you new assistant name to navigate to the **AiseraGPT Details** Window. 2. Select one of the Channels you added by double-clicking on the channel name.

3. Click the **Test** button in the lower-left corner of the **Channel Details** Window.

4. You will now be able to to see a test version of you assistant.

## To Brand Your Virtual Assistant: 1. Go into **Settings > Channels**, in the Aisera Administration application, and select your virtual assistant. 2. In the **Channel Details** window, click the **Pencil** icon to start editing the channel.

3. On the **Edit** **Channel** screen navigate to the **Styling** tab. there's a place to add your logo, choose colors for different parts of your virtual assistant screen, set window size, add timeout parameters, add footers, and privacy messages and options. 4. You will be able to edit the logo, colors, name, opening message and more from this window. Edit the channel to fit the desired style and select **OK** to save the changes.

## Testing Your Bot Outside the Admin UI Copy the JavaScript snippet located at the bottom of the **Channel Details** window. You can use the JavaScript Snippet to test your Bot in a browser window (with Chrome Developer Tools).

# How to Create a Conversational AI 2.0 Bot ## Create a Conversational AI 2.0 Bot To create a Conversational AI 2.0 bot: 1. Navigate to **Settings > AiseraGPT** and click on + **New AiseraGPT** button and add the mandatory details in the window 2. Select the **Advanced** tab 3. Toggle **Conversational AI v2.0** > **Enable** 4. Click **OK** to create the new AiseraGPT. Once a bot has been create you will not be able to change the type. You will need to create a new AiseraGPT if you want to use the Conversation 1.0 bot. ## Try 2.0 Workflow Search With a Test Bot You can ask your Aisera team to help you set up a **Conversational AI 2.0** workflow at the bot level, instead of the tenant level, to test out the search and fulfillment capabilities. If you want to try this option, ask your Aisera CSM to enable the appropriate feature flags. ## Ability to Make Time-Related Requests When your bot has Conversational AI 2.0 enabled, your bot users can make relative time-related Requests such as: * What are the titles of the Knowledge Base Articles added in the past 40 days? * Who won the Outstanding Team Award in the last quarter of 2025? * How many shopping days are there until Christmas? * Will there be any public holidays in the next 10 days? * Do we have a long weekend at the end of the month?
# Embed Your Webchat Channel

This topic describes the installation of the Aisera Webchat bot (AiseraGPT application on the Aisera Webchat channel) on any portal or web landing page. This involves setting up session variables to pass information to the bot before loading it, and also the specific JavaScript code that should be used in the page layout to load the bot on the page. ## JavaScript for Aisera Webchat Bots The Aisera Webchat Bot is included as a default channel when you create a new application/bot. \ You can use the channel 'as is', or add different channels, for a different 'look and feel'.

If you're using the default Webchat channel, you can add it to an HTML web page by including the parameters within a JavaScript( JS) tag. on your page or portal. \ \ The following is an example Webchat script. {% code overflow="wrap" %} ```javascript (function() {var d = document, s = d.createElement('script'); s.async = true; s.src = 'https://.chatbot.aisera./awc/js?t=a1b2c3d4-e5f6-7890-1234-567890abcdef' + Date.now(); d.getElementsByTagName('head')[0].appendChild(s)})(); ``` {% endcode %} #### To obtain the Webchat Script: 1. In the Aisera Admin UI navigate to **Settings > AiseraGPT** in the left navigation panel. 2. In the **AiseraGPT Details** window under **Channels** click on the **Webchat** channel. 3. At the bottom of the **Channel Details Page** under the **JavaScript** **Snippet** click **Copy.**

4. This script contains all the relevant parameters for loading your bot on your page or portal. \ To write JavaScript code inside an HTML page, you must enclose your code within **`` tags**. Add your webchat javascript (like thie one in the example above) between two script tags, as shown in the red box in the example below.

**NOTE:** If you choose the **Test** button below the JavaScript snippet (shown in the diagram as part of Step 3 above, the script will render a Webchat bot, according to the JavaScript instructions.

## Additional Configurations The following sections contain additional options, but you must perform these edits and operations outside the regular commands in the Aisera Admin UI. ### Session Variables Session variables can optionally be passed to Webchat in the `window.aisera_bot` global variable (or `window.webchat` for backward compatibility). The following session variables have special meaning to Webchat: * `userEmail:` User email * `userFullName:` User full name In addition to the user variables, other session variables could be specified to be used in custom flows. For example: ```javascript window.aisera_bot = { userEmail: "jsmith@acme.com", userFullName: "John Smith", accountId: "1234" } ``` {% hint style="info" %} If you specify `userEmail` in the session variables then you have to set the variable before the snippet that loads Webchat. *It cannot be changed later* using the [update session vars](#update-session-vars) message. {% endhint %} Below is an example showing session variables being passed to the webchat. ```html

Sample page

``` ### Input Messages You can make a request to Webchat to perform operations using the `window.postMessage` function. The JavaScript for these messages looks like the following: ```javascript let message = { type: MESSAGE_TYPE, // additional fields depending on the message type }; window.postMessage(JSON.stringify(message), "*"); ``` The following input message types are supported: #### Open This message opens the application window if it is not already opened.

Field	Description
type	`aisera.webchat.open`
text	Optional text to be used as a search utterance
message	Optional html to be displayed in Webchat Other fields can be specified here and received in the opened output messages described below.

#### Close This message closes the webchat window if it is not already closed

Field	Description
type	`aisera.webchat.close`

#### Reset This message clears the webchat content and the current conversation contexts.

Field	Description
type	`aisera.webchat.reset`

#### Hide This message hides the Webchat and the opener icon.

Field	Description
type	`aisera.webchat.reset`

#### Show This message shows the Webchat and the opener icon if they are hidden using the `aisera.webchat.hide` message.

Field	Description
type	`aisera.webchat.show`

#### Update Session Vars The hosting page can update the `window.aisera_bot_variable` at any time, but it needs to send this notification to notify the application.

Field	Description
type	`aisera.webchat.update_session_vars`

### Output Messages Webchat sends notification messages to the host window during specific operations. The JavaScript that allows Webchat to receive the messages looks like the following: ```javascript window.addEventListener("message", event => { let message = JSON.parse(event.data); // use message.type etc }) ``` The following outgoing message types are supported: #### Ready This message is sent when Webchat is initialized and is ready to process messages.

Field	Description
type	`aisera.webchat.ready`

#### Opened This message is sent when Webchat is opened. All fields specified in the [open](#open) input message are included in the message.

Field	Description
type	`aisera.webchat.opened`

#### Closed This message is sent when Webchat is closed.

Field	Description
type	`aisera.webchat.closed`

#### Handoff This message is sent when a button with the Hand-off action is clicked.

Field	Description
type	`aisera.webchat.handoff`
transcript	The conversation transcript as a `string`

#### Session Timeout This message is sent when the conversation session times out. The session times out when the user idle timeout configured in the Admin UI elapses.

Field	Description
type	`aisera.webchat.session_timeout`

## Webchat Embed with an Iframe To use your Webchat in an `iframe` inside a page containing a webchat snippet, change the URL in the snippet to include a `target_id` parameter with the `iframe tag id`. For example, if the page has the following tag: ```html