Using Amazon Quick Flows - Amazon Quick Suite
What are Flows?Why Flows?Quick Flows features and capabilitiesCollaboration and governance

Using Amazon Quick Flows

Amazon Quick Flows is a capability within Amazon Quick Suite that empowers business users to transform their everyday tasks into workflows for individual and team productivity. It enables any user to create, customize, and share purpose-built workflows utilizing their data, insights, and actions available within Amazon Quick Suite. These intelligent flows can be generated from conversations with chat agents or by describing requirements in natural language, without requiring any technical skills. They can also be published to an admin-managed library and shared with other Amazon Quick Suite users in the organization.

Creators can describe a pre-defined set of steps that accomplish tasks through a predictable, reusable flow. These flows can fetch information, take action, generate content, and handle process-specific requirements through creator-specified logic.

Building on top of Amazon Quick Suite - that brings together their data, analysis, and actions securely in one place, business users can describe the repetitive tasks they need automated. With Amazon Quick Flows, business users can streamline their routine tasks like generate responses for RFPs, review statements of work (SOW), or collate the latest industry trends and insights into a sales pitch.

Amazon Quick Flows offers agentic runtime support enabling end users to chat with their workflow and request updates (like summarizing an output or skipping a step). The building experience offers simple patterns to generate responses from various sources of data, and allows creators to specify instructions and "logic" within a flow. Creators can also add the ability to repeat a subset of steps, apply if-then-else logic, validate input before generating output and so on, all through simple, natural language instructions. Amazon Quick Flows offers expanded data options to generate AI-powered responses from Amazon Quick Sight to spaces to latest updates from web search. Amazon Quick Flows empower users to process multi-modal inputs (including images and video) and refine their outputs for 'faster' or more 'balanced and versatile' responses. With the UI agent feature in Amazon Quick Flows, users can navigate public websites (that don't require a login) and perform tasks (like scroll to find information, fill forms). All this while giving IT teams peace of mind with governance controls to manage access to select features (like web search or UI agent) and gate the sharing of flows across the organization by mandating approval reviews.

What are Flows?

Amazon Quick Flows is a capability within Amazon Quick Suite that empowers business users to transform their everyday tasks into workflows for individual and team productivity. Quick Flows makes it easy to design and manage workflows through a no-code interface, so you can build workflows without writing any code.

Using Amazon Quick Flows, you can create workflows that combine processes using AI-generated responses from their data in Quick suite, web search, direct model responses or generate images. You can write agent instructions to perform if/else statements, loops and conditions and leverage UI actions to interact with websites. You can also use integrations to connect to their external services and gather User Input as text or uploaded files.

When running a flow, users get the option of interacting with their workflows as a sequence of pre-defined steps, along with the flexibility of conversational UI to chat with it to refine outputs or ask follow-up questions. Each flow consists of the following steps:

AI-generated responses

  • Quick Suite data step that retrieves responses from spaces and knowledge bases.

  • Quick Sight step that gets insights from your dashboards and topics.

  • Web step that generates results from internet search.

  • General knowledge step that generates responses directly from models

  • Image steps to generate AI images from text and image inputs.

Instruct agent

  • Reasoning group step that process if, when, and how to run steps.

  • UI agent step to perform tasks on public websites.

Integrations

  • Action step that performs read or write operations in connected systems

User input

  • Text step to gather free-form text input from users.

  • Files step to accept files from the user.

Why Flows?

Organizations today face increasingly complex business processes that often require both human judgment and system interactions. Amazon Quick Flows bridges this gap by combining AI reasoning capabilities with direct business actions.

Consider a typical customer support scenario where a customer submits a technical issue. Without automation, this process involves multiple manual steps: reading the ticket, researching the issue, consulting knowledge bases, possibly escalating to specialists, and finally responding with a solution.

With Amazon Quick Flows, business users can streamline their routine tasks like generate responses for RFPs, review statements of work (SOW), or collate the latest industry trends and insights into a sales pitch. You can transform this process through:

  1. Automated information gathering: Input steps collect and process customer information when a support ticket arrives.

  2. Intelligent analysis: Amazon Bedrock models analyze the ticket content and identify potential solutions.

  3. Seamless system integration: Action steps connect to your knowledge bases, CRM systems, and other tools to gather context and update records.

  4. Contextual decision making: Reasoning defines how and when the flow runs, guiding multiple steps such as applying conditions, performing validations, or running loops to determine the appropiate next step.

  5. Flexible output delivery: Output steps provide solutions to customers in the appropriate format, whether through chat interfaces or structured responses.

Creators can describe a pre-defined set of steps that accomplish tasks through a predictable, reusable workflow. These workflows can fetch information, take action, generate content, and handle process-specific requirements through creator-specified logic.

Amazon Quick Flows offers agentic runtime support enabling end users to chat with their workflow and request updates (like summarizing an output or skipping a step). This flow reduces response times from hours to minutes while ensuring consistent quality and 24/7 availability. As your business grows, Quick Flows scales effortlessly, handling increasing volumes without proportional increases in staffing.

Beyond customer support, Amazon Quick Flows can transform numerous business processes across departments:

  • Sales, marketing, and operations: Qualifying leads, generating personalized proposals, creating marketing content, updating CRM records, and supporting processes like RFP responses

  • HR: Processing employee requests, answering policy questions, and automating onboarding steps

  • Finance: Analyzing expense reports, flagging anomalies, and processing routine approvals

  • IT: Automating troubleshooting, system monitoring, and access management

By combining AI reasoning with direct business actions, Amazon Quick Flows unlocks new levels of efficiency and scalability across your organization. For more detailed information about how Flows work, see Terminology and key concepts.

Quick Flows features and capabilities

Enterprise customers can use Amazon Quick Flows, a feature within Amazon Quick Suite, to leverage intelligent workflow capabilities. Quick Flows allows business users to predefine a set of steps across the applications and services they use to get work done, to save time and effort on repetitive business processes. Creators can build Quick Flows by providing instructions on the user defined goal in natural language. Quick Flows will apply Quick Suite level governance like guardrails and determining permissions at a user level to restrict access to data sources and tools. Quick Flows provides an intuitive visual creation experience building experience with visual flow design and step configuration.

Quick Flows runtime provides on-demand execution where users can trigger flows interactively. Runtime chat mode enables conversational experiences within Flows, while runtime structured mode provides guided interfaces for standardized processes. Creators can configure core capabilities like name, description, and set instructions for how and when steps run (reasoning) for the flow and define a sequence of steps that generate AI-powered responses from data across Quick Suite and the web, take actions in third party applications, and apply agentic reasoning to control execution of steps or even navigate websites with human-like interactions. Creators can share their Quick Flows with individuals or groups as either co-owners (who can edit and manage the flow) or viewers (who can only run the flow), or with all users who have appropriate access permissions.

Quick Flows offers a comprehensive set of features designed to streamline your repetitive tasks. Here's what you can expect:

Administration

  • User Roles integrated with Amazon Quick Suite identity management

  • Ability to enable/disable flows for the Quick account

  • Ability to enable/disable usage of Bedrock models for output refinement in Flows

  • Ability to enable/disable access to internet to enhance responses

  • Ability to enable/disable access to UI agent to perform browser tasks

Creation

  • Use natural language prompts (NLP) to describe the flow to create

  • Create Quick Flows manually using the visual UI

  • Transform conversation with chat agent into a flow

  • Duplicate shared Quick Flows

  • Step-level editing capabilities

  • Different step types supported including file upload (default file) and text input (static text)

  • Reasoning instructions (loops, validations, conditions) to specify how one or more steps should execute within a group

  • Leverage General knowledge (instead of Company knowledge) (powered by Amazon Bedrock) for text and image output steps with multi-modal inputs (including images and video)

  • Actions to perform read/ write operations connected applications

  • Amazon Quick Sight step that generates response from dashboards and topics

  • Web search to enhance responses with latest updates from the internet

  • UI agent (Preview) to navigate public websites (that don't require a login) and perform tasks

Runtime

  • Conversational runtime support for follow-up questions, refining outputs, and error state handling

  • Structured runtime with steps

  • Agentic runtime validates inputs based on constraints defined in reasoning instructions

  • Visual indication of flow execution progress

  • Support to view and resume past flow runs through history

Collaboration and governance

Amazon Quick Flows includes comprehensive features for team collaboration and organizational governance, giving IT teams peace of mind with governance controls to manage access to select features (like web search or UI agent) and gate the sharing of flows across the organization by mandating approval reviews.

Sharing controls

Quick Flows provides flexible sharing options to enable collaboration while maintaining organizational control:

  • Share with individual users, teams, or with all users using the Quick account

  • Share access can be as viewer enabling other users to run the flow or as co-owners, allowing them to edit and further share a given flow

  • Ability to unpublish any shared flows and manage ownership transfer of flows

Approval workflows

Administrators can implement approval processes to ensure proper governance before flows are shared across the organization:

  • Ability to enable/disable need for approval review before sharing