Skip to main content

Overview

Supatest’s AI capabilities transform how you create tests. The AI workflow uses two agents working together: the Test Planner Agent for planning what to test, and the AI Action Agent for automating test steps.

How AI Test Creation Works

1. Test Planner Agent (Global Chat)
   ↓ Creates test plan with scenario notes

2. Review and refine the plan
   ↓ Edit scenario notes as needed

3. AI Action Agent (Inside Editor)
   ↓ Automates scenario notes into steps

4. Review and verify
   ↓ Add assertions manually

AI-Powered Tools

Test Planner Agent

The Test Planner Agent is the global chat that helps you plan what to test. It generates test cases with scenario notes that describe the test flow.

How It Works

  1. Open the global AI chat
  2. Describe what you want to test
  3. Planner creates test cases with scenario notes
  4. Review and edit the scenario notes
  5. Use AI Action Agent to automate

Example

You describe:
Test the checkout process for an e-commerce site
Planner generates test cases with scenario notes: Test Case 1: “Successful checkout”
1. Navigate to products page
2. Add first product to cart
3. Go to checkout
4. Fill shipping information
5. Enter valid payment details
6. Complete purchase
7. Verify order confirmation
Test Case 2: “Invalid payment handling”
1. Add product to cart
2. Go to checkout
3. Fill shipping information
4. Enter invalid credit card
5. Submit payment
6. Verify error message appears

Best For

  • Planning multiple test cases at once
  • Generating edge cases you might miss
  • Creating test scenarios from requirements
  • Building comprehensive test suites
Learn more about Test Planner Agent

AI Action Agent

The AI Action Agent lives inside the test editor. It takes scenario notes and automates them into executable test steps.

How It Works

  1. Open a test case in the editor
  2. Review the scenario notes
  3. Click to run AI Action Agent
  4. Agent performs actions based on scenario notes
  5. Steps are created as it executes

What AI Action Agent Can Do

Action TypeExamples
ClickButtons, links, checkboxes, radio buttons
FillText inputs, textareas, search fields
SelectDropdowns, option lists
ScrollPage scrolling, element scrolling
NavigateFollowing links, URL changes

What AI Action Agent Cannot Do

AI Action Agent focuses on primary interactions. It cannot add assertions or verification steps. You must add these manually using the recorder or visual editor.
Requires manual addition:
  • Verify Text steps
  • Verify Visibility steps
  • Verify URL steps
  • Visual assertions
  • Any verification/assertion logic

Example Workflow

Scenario Note: "Login with test credentials"

AI Action Agent performs:
→ Navigates to login page
→ Fills email field
→ Fills password field
→ Clicks submit button

You add manually:
→ Verify Text: "Welcome, User"
→ Verify URL contains: "/dashboard"
Learn more about AI Action Agent

Combining Planner and Action Agent

Complete Workflow

  1. Plan with Planner Agent
    • Open global chat
    • Describe your testing goals
    • Get test cases with scenario notes
  2. Automate with Action Agent
    • Open test case in editor
    • Run Action Agent on scenario notes
    • Watch as steps are created
  3. Add Assertions Manually
    • Use recorder to capture verification elements
    • Add Verify Text, Verify Visibility steps
    • Configure expected values
  4. Review and Refine
    • Run the complete test
    • Fix any failing steps
    • Adjust locators if needed

Example: Building a Login Test Suite

Step 1: Planner creates tests
You: "Create tests for the login page"

Planner generates:
- Test 1: Valid login succeeds
- Test 2: Invalid password shows error
- Test 3: Empty fields show validation
Step 2: Action Agent automates each test
Test 1 scenario notes:
1. Navigate to /login
2. Enter valid email
3. Enter valid password
4. Click login button

Action Agent creates steps:
→ Navigate: /login
→ Fill: email input
→ Fill: password input
→ Click: submit button
Step 3: You add assertions
Manually add:
→ Verify Text: "Welcome back"
→ Verify URL: /dashboard

AI Credit Usage

AI features consume credits from your account:
FeatureCredits
Test Planner generation1-3 credits per plan
AI Action Agent execution1 credit per action
Learn about managing AI credits

Tips for Better Results

Write Clear Scenario Notes

Less effective:
"Test login"

More effective:
"Navigate to login page, enter [email protected] as email,
enter password123 as password, click the login button"

Be Specific About Elements

Include details that help AI find the right elements:
  • Button text: “Click the ‘Add to Cart’ button”
  • Input labels: “Fill the ‘Email Address’ field”
  • Location hints: “Click submit button at the bottom”

Always Add Assertions

Never rely solely on AI Action Agent. Every test needs manual assertions to verify the expected outcome.

Next Steps