Announcing embedded chat in Amazon Quick Suite

Today, we’re excited to introduce Amazon Quick Suite embedded chat, a unified conversational experience you can embed directly into your own applications. With this launch, you can bring agentic AI-powered chat from Quick Suite, which unifies both structured data and unstructured knowledge in a single conversation, into the tools where users already work. This makes it straightforward for organizations to add intelligent, contextual answers inside their applications without building conversational interfaces, orchestration logic, or data access layers from scratch.

As organizations adopt AI, one pattern is clear: people don’t want to switch tools to get answers. They want to ask a question in the moment—in their CRM, support console, analytics portal, or internal dashboard—and get a precise, contextual response. Similarly, independent software vendors (ISVs) want to integrate advanced agentic capabilities into their customer-facing products. However, most conversational tools still force a compromise: they are either great at structured data or great at documents, more suitable for analytics or good for knowledge bases, and able to answer questions or perform actions. They are rarely all of the above.

The unified chat feature in Quick Suite addresses this need with agentic AI that can reason over dashboards, documents, indexes, and connected data in a single conversation. A user can reference a KPI, pull details from a file, bring in customer feedback from a ticket, and execute a defined action, all without leaving the chat interface in their tool of choice. With unified embedded chat, organizations can now place this agentic AI-driven experience directly inside their own products and portals, giving users a powerful, personalized experience that fits naturally into their workflow. Quick Suite empowers users to launch and embed an intelligent conversational agent in just a few minutes, with integrations to 40+ data sources.

In this post, we’ll help you unlock the power of Quick Suite’s embedded unified chat, bringing AI-powered assistance directly into your workflows.

Key features and benefits

The embedded unified chat feature offers the following key benefits:

Unified chat experience across structured and unstructured data – The embedded chat brings the agentic reasoning of Quick Suite—across dashboards, files, notes, and connected sources—into your applications using a wide variety of data. Users can ask natural-language questions, explore summaries, compare insights, and run available actions, all in one continuous flow working seamlessly across structured and unstructured data sources.
Connectivity to your favorite applications – Quick Suite brings your team’s applications directly into the conversation. Through these connections, users can search through documents stored in SharePoint and OneDrive, send messages in Slack, create tasks in Jira, or use custom connections enabled through MCP, all without leaving chat. Your users get instant access to the information and actions they need, right where they’re already working.
One-click embedding – With Quick Suite, you can quickly embed a chat agent into an application within minutes by simply copying the code from Quick Suite interface and inserting it into your enterprise application.
Security and access controls – Embedded chat is secure by default. Data powering the embedded conversational experience stays under your control. You choose what the agent can access—a curated Space, an existing Q index, or connected data sources. Actions are also explicitly scoped, giving teams full governance while still benefiting from Quick Suite agentic AI.
Customizable visual theming to match your company’s brand – Quick Suite offers powerful customization capabilities that extend your organization’s brand identity directly into the embedded chat experience. You can customize Quick Suite with your company’s brand colors, creating a cohesive visual identity across your entire platform. Because you can integrate the Quick Suite chat functionality into your application using an iframe, the widget seamlessly inherits your brand’s look and feel with no additional configuration required.
Customizable tone to match your company’s voice – Quick Suite lets you infuse your company’s unique voice into every interaction. You can create a custom chat agent with tailored instructions that reflect your company’s communication style and expertise and set the tone, whether professional and formal, friendly and conversational, or technical and precise. Quick Suite lets you welcome users with personalized welcome messages that align with your brand identity, so conversations start on the right note. You can also give it specific instructions on how to format answers, or give it hints about answering questions unique to your use cases.

This comprehensive customization, from visual theming to conversational tone to application integrations, makes sure the embedded chat widget blends naturally into your existing applications, providing users with a consistent, branded experience.

You can use the Quick Suite embedded chat feature at no additional cost, at existing Quick Suite pricing.

Solution overview

To embed the unified chat, you must use the GenerateEmbedUrlForRegisteredUser or GenerateEmbedUrlForRegisteredUserWithIdentity API. To make it visible to users in your application, you must use the Quick Sight Embedding SDK version 2.11.0 or higher.

Prerequisites

Make sure you have the following prerequisites:

An AWS account set up with Quick Suite enabled. If you don’t have a Quick Suite account, you can sign up for one. For instructions to get started with a Quick Suite free trial, refer to Getting started with Amazon Quick Suite.
An AWS Identity and Access Management (IAM) role with associated policies to provide permissions to users who assume it. The following is a sample policy statement:

{
    "Action": [
        "quicksight:GenerateEmbedUrlForRegisteredUser"
    ],
    "Resource": "*",
    "Effect": "Allow"
}

Put the domain where you want to embed the dashboard on your allow list. This can be done on the Quick Suite console’s Manage Amazon Quick Suite menu, under Security / Manage domains.
The unified chat agent embedding is currently available for registered users with either the professional or enterprise licensing. Make sure you have the user provisioned in Quick Suite.

Create a chat agent

To embed a chat agent, first create and configure a custom chat agent in your Quick Suite application. By building a custom chat agent, you define the agent’s persona using natural languages: who the agent is (identity and role), what the agent does (core responsibilities), and how the agent communicates (tone and style). You can further configure your chat agent’s knowledge sources (including dashboards, topics, and unstructured data), as well as actions and integrations to help the agent take further steps as an integral part of the chat experience (such as Slack, Outlook, or Jira). For instructions on creating a chat agent, see Create, customize, and deploy AI-powered chat agents in Amazon Quick Suite.

One-click enterprise embedding

With one-click embedding, you can embed a Quick Suite unified chat agent with a static embed code from Quick Suite that is added to an iframe. When a user accesses the chat agent on your intranet enterprise applications, they will be required to sign in to Quick Suite. The following screenshot illustrates how to retrieve the embed code.

Registered user embedding

You can also embed the unified chat agent using registered user embedding. With registered user embedding, you can seamlessly integrate Quick Suite chat agents in your custom application while using your organization’s existing enterprise authentication infrastructure, so users authenticate one time through your corporate identity provider and gain access to the chat agent without additional login prompts. In addition, each user receives a personalized conversational experience where they see only the information they’re authorized to access, all while maintaining the robust security standards your enterprise requires.

Complete the following steps in this section to embed the unified chat agent using an API.

Generate secure embed URL

Generate and extract a secure embed URL using the GenerateEmbedUrlForRegisteredUser or GenerateEmbedUrlForRegisteredUserWithIdentity Quick Suite API. The following is a sample Python code snippet. For a detailed example, refer to the SDK documentation.

# Generate embed URL for QuickChat experience
response = quicksight_client.generate_embed_url_for_registered_user(
    AwsAccountId='<ACCOUNT_ID>',           # Your 12-digit AWS account ID
    UserArn='<USER_ARN>',                  # ARN of the registered Quick Sight user
    ExperienceConfiguration={
        'QuickChat': {}                    # Specifies QuickChat embedding experience
    },
    AllowedDomains=[
        '<URL>',                           # Domain(s) where embedding is allowed
    ]
)

# Extract the embed URL from the response
embed_url = response['EmbedUrl']

The ExperienceConfiguration parameter with 'QuickChat': {} specifies you want to embed the Quick Suite chat agent interface.

Configure chat interface

Use the QuickSight Embedding SDK to render the chat interface in your web application. Use the following key configuration options:

frameOptions.url – Provide the secure embed URL generated in the previous step.
frameOptions.container – Provide the CSS selector for the DOM element to contain the chat.
contentOptions.fixedAgentArn – Provide an optional Amazon Resource Name (ARN) to embed a specific custom chat agent. The default is the Quick Suite system chat agent.

To get the custom chat agent ARN, complete the following steps:

On the Quick Suite console, choose Explore in the navigation pane, then choose Chat agents.
Under the Action column, choose the options menu next to Chat and choose View chat agent details.
Choose Copy link next to the chat agent name.

The link will look similar to the following URL:

https://us-east-1.quicksight.aws.amazon.com/sn/start/agents?view=6fxxxxxx-xxxx-xxxx-xxxx-xxxxxx43137d

The text after view= is the agent ID.

The agent ARN is formatted as arn:aws:quicksight:us-east-1:<<aws-account-id>>:agent/<<agent-id>>.

The following code sample shows the frameOptions and contentOptions parameters used to load the unified chat experience:

// Configure the iframe container and dimensions
const frameOptions = {
    url: "<YOUR_EMBED_URL>",        // URL from Step 1 backend call
    container: '#experience-container',
    height: "700px",
    width: "1000px",
    onChange: (changeEvent, metadata) => {
        switch (changeEvent.eventName) {
            case 'FRAME_MOUNTED': {
                console.log("QuickChat frame successfully mounted");
                break;
            }
            case 'FRAME_LOADED': {
                console.log("QuickChat experience loaded and ready");
                break;
            }
        }
    },
};

// Configure chat-specific options
const contentOptions = {
    fixedAgentArn: '<AGENT_ARN>',   // Optional: specify the agent Arn to embed
    onMessage: async (messageEvent, experienceMetadata) => {
        switch (messageEvent.eventName) {
            case 'CONTENT_LOADED': {
                console.log("QuickChat interface initialized successfully");
                break;
            }
        }
    }
};

Real-world use cases

Real-world application: Financial analysis assistant embedded in a finance performance dashboard

A fictitious company embeds a custom chat agent, a financial analysis assistant with knowledge of the underlying finance dataset and business context documents, into their finance performance dashboard. The dashboard is accessed by hundreds of diverse users across executive leadership, their internal finance team, and business leads, including product owners, regional managers, and sales leaders.

Democratized data access and self-service analytics

Because the chat agent is linked to the underlying dataset using the topic setup, non-technical users can explore data independently. Users who aren’t comfortable with filters or parameters can ask natural language questions such as “What are sales figures for the EMEA region,” “What was our revenue in Q3 2024,” and “Which country had the highest revenue growth last quarter?”

Business users routinely have follow-up requests as they interact with a self-service dashboard. With the help of the chat agent, they can now automate the generation of additional data queries or reports without relying on the BI team. For example, they can streamline routine reporting tasks by having the agent generate the following summaries:

“Create a monthly financial summary for the executive team highlighting key metrics and trends”
“Compare our performance against budget targets across all departments”
“Show me profit margins by product line for the current quarter”

From data to real-time insights: Understanding the “how” and “why”

Furthermore, by embedding an agent that combines structured data from the dashboard with unstructured data from company documents (board reports, strategy memos, market analyses), end-users can ask questions about the “how” and “why” behind the data in real time without leaving the dashboard or switching applications. While viewing a revenue decline chart, they can immediately ask “What caused this drop?” and get answers that combine the visible data with underlying documents. Some example questions they might pose include:

“Why did revenue decline 15% in Q3 compared to Q2?”
“I see Product A’s profit margin dropped 8% this quarter. What did the product strategy team recommend in their last quarterly review?”
“We’re showing $45k in software licenses this month. How should this be classified according to our accounting policy manual?”

From insights to real-time action: Integrations that drive decisions

As the embedded chat agent integrates with external applications such as emails and ticketing systems, it transforms the finance dashboard from a passive reporting tool into an active command center that can initiate and track actions directly in the team’s workflow system. Some example use cases include:

Alerting stakeholders through Slack – After noticing an anomaly in Q4 operating expenses, a finance analyst can ask the chat agent to “Send a slack to the finance directors summarizing this expense spike with a link to this dashboard.”
Creating follow-ups in Asana – Directly from a dashboard insight, a regional manager can say “Create an Asana task for the country lead to review this revenue decline. Assign it high priority with a due date of next Friday.”

Upcoming features

In 2026, we plan to broaden how customers can use embedded chat, including more ways to tailor the experience, reach new audiences, and integrate deeper into hosted applications. This might include expanded custom branding controls, support for unauthenticated (anonymous) users, and additional embedding surfaces. We’re also exploring richer internationalization and more flexible deployment patterns based on customer feedback.We will share more as these capabilities come to life.

Conclusion

The launch of Quick Suite embedded chat represents a significant leap forward in making AI-powered conversations more accessible and integrated within enterprise workflows. This launch addresses the fundamental challenge organizations face today: enabling users to access powerful AI capabilities without leaving their familiar work environments. This solution solves this challenge of tool fragmentation while providing enterprise-grade security and extensive customization options that adapt to your brand’s look and feel. Quick Suite embedded chat delivers a comprehensive solution that grows with your needs, such as exploring data through natural language queries, connecting insights across structured and unstructured sources, or triggering workflow actions. As we continue to enhance the platform with upcoming features like expanded branding controls and support for anonymous users, we’re excited to see how organizations will transform their user experiences with this technology.

For more detailed information about the Quick Suite SDK and experience-specific options, visit the GitHub repo.

Join the Quick Suite Community to ask, answer, and learn with others and explore additional resources.

About the authors

Salim Khan is a Specialist Solutions Architect for Amazon Quick Suite. Salim has over 16 years of experience implementing enterprise business intelligence (BI) solutions. Prior to AWS, Salim worked as a BI consultant catering to industry verticals like Automotive, Healthcare, Entertainment, Consumer, Publishing and Financial Services. He has delivered business intelligence, data warehousing, data integration and master data management solutions across enterprises.

Pallavi Sharma is a Principal Product Manager for Amazon Quick Suite, where she leads embedding across the Quick Suite portfolio. She has over a decade of experience building and scaling enterprise software across cloud modernization and management tools, low-code platforms, and AI-driven solutions. An electrical engineer by training, she began her career as an ASIC design engineer in the semiconductor industry. Pallavi is passionate about creating intuitive, human-centered products that help organizations unlock the power of AI and simplify how people work.

Marisa Parker is a Sr. UX Design Manager for Amazon Quick Suite, where she leads a team of designers building enterprise AI experiences. With over a decade of experience, she has shaped AI experiences across the technology industry, including leading design teams for developer tooling, cloud platforms, and enterprise resource planning systems. She is passionate about creating intuitive experiences that help users use AI to enhance productivity and streamline workflows.

Joy Cheng is a Senior Builder at AWS, focusing on Amazon Quick Suite implementations for enterprise clients across industries. She brings over a decade of experience in enterprise data solutions, having held analytics engineering and leadership positions at Amazon Music and NBCUniversal, with her consulting portfolio extending to education and health initiatives for the US government and ministries across Africa and Asia. A former instructor for data analytics bootcamps at UC Irvine and UCSD, she is passionate about making AI and data analytics accessible to everyone.

AWS Business Intelligence Blog