In-flow user input
Inline interactions that collect structured input or allow users to review and modify agent-generated output within a generative AI chat.
Inline interactions that collect structured input or allow users to review and modify agent-generated output within a generative AI chat.
Only present an interaction when the agent understands the user's intent and needs something specific from the user to proceed. Don't front-load interactions that could be deferred or resolved automatically.
Help users understand the value in providing additional information. Highlight how their input enables the agent to generate better output or execute actions correctly, and the expected outcome.
Display input methods that complement the type of information to be collected. Questions with a finite set of answers, such as selecting resource IDs or timestamps, should use a succinct form. This reduces errors and improves data quality.
An inline form collects the required values directly in the conversation. For example, agent is helping create a resource and progressively displays forms inline based on what information is needed. The steps are synchronous; the output of the previous step, determines the next. Forms appear sequentially and may have agent responses between each step. Completed steps show a summarized read only view.
I'll create a Lambda function triggered by an EventBridge schedule rule to run your cleanup job every night. I just need a few configuration details.
Steps are independent and the agent can process them without waiting for previous results. Forms appear as individual collapsible cards within the conversation. Each step resolves on its own and doesn't block subsequent steps, similar to asynchronous creation.
Happy to help. I'll walk you through the setup step by step.
An inline edit within the response allows the user to review and optionally modify the value the agent has created.
I've created an environment for you so we have a place to track everything, code, infrastructure, and deployments.
Now a few quick questions so I can set up the right backend for you:
Use sentence case, but continue to capitalize proper nouns and brand names correctly in context.
Use end punctuation, except in headers and buttons. Don’t use exclamation points.
Use present-tense verbs and active voice.
Don't use please, thank you, ellipsis (...), ampersand (&), e.g., i.e., or etc. in writing.
Avoid directional language.
For example: use previous not above, use following not below.
Use device-independent language.
For example: use choose or select not click.
Be specific to the resource or action. Avoid generic labels like Information required or Configuration needed.
Follow the guidelines on alternative text and Accessible Rich Internet Applications (ARIA) regions for each component.
Make sure to define ARIA labels aligned with the language context of your application.
Don't add unnecessary markup for roles and landmarks. Follow the guidelines for each component.
Provide keyboard functionality to all available content in a logical and predictable order. The flow of information should make sense.