> ## Documentation Index
> Fetch the complete documentation index at: https://docs.stably.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Action

> Perform complex multi-step actions in plain English with Stably's AI agent.

<Note>
  This documentation is for **Stably Classic**. For the new Stably Web Editor, see [AI Agent Execute](/stably-sdk/ai-agent-execute).
</Note>

Stably is the only testing platform that allows you to perform complex multi-step actions in plain English.

Not only do we have the best-in-class AI agent, we also build a knowledge base for each app to ensure the AI can handle your unique business logic.

<iframe
  src="https://www.loom.com/embed/92b19b705dae4636977bd03e51446803?sid=dafaa137-eb07-4778-ba27-4b2b170c15da"
  style={{
width: '100%',
aspectRatio: '16/9',
border: 'none',
borderRadius: '8px'
}}
/>

## How AI Actions Can Help

AI actions enhance your testing workflows by solving complex scenarios. Here are five concrete examples of how they're used:

1. **Test Data Reset** - Reset application data to a specific state at the beginning of tests with commands like "Clear all user data" or "Reset inventory to default values."

2. **Dynamic Popup Handling** - Automatically detect and respond to unexpected popups, alerts, or dialogs with instructions like "Dismiss any error popups that appear" or "Accept cookies if prompted."

3. **Conditional Navigation** - Navigate through different paths based on application state with steps like "If the dashboard shows zero orders, create a new order, otherwise view the latest order."
   <Tip>You can also use [conditional actions](/platform-capabilities/conditionals) to do this.</Tip>

4. **Environment Verification** - Verify the testing environment is properly configured before test execution with commands like "Check if the test database is connected" or "Verify the API endpoints are accessible."

5. **Complex Data Input** - Handle complex data entry scenarios by instructing the AI to "Fill the form with valid user information" or "Generate and input random test data that meets validation requirements."

## Add AI Action to your test

You can add an AI Action to your test by clicking the "AI Action" button in the test editor.

<Frame type="glass">
  <img src="https://mintcdn.com/stablyai/kbD1mPIS9Ksvjr53/images/no-code-editor/manual-add-ai-action.png?fit=max&auto=format&n=kbD1mPIS9Ksvjr53&q=85&s=ad71c5d3d50b515cac1c47c43348b8d6" width="342" height="461" data-path="images/no-code-editor/manual-add-ai-action.png" />
</Frame>

You can also convert multiple Playwright actions to an AI Action by clicking the "AI Action" button in the action list.

<Frame type="glass">
  <img src="https://mintcdn.com/stablyai/kbD1mPIS9Ksvjr53/images/no-code-editor/convert-to-ai-action.gif?s=b1509c8b2ae703b1de5ece23b8f84448" width="370" height="520" data-path="images/no-code-editor/convert-to-ai-action.gif" />
</Frame>
