FAQ

What is Panto AI QA?

Panto AI is an end to end mobile QA testing platform. You describe a feature in natural language, the AI Agent executes the flow on the app, and the system generates deterministic Appium or Maestro scripts that can run on any device, any version, any time through integrated device farms.

How do I connect my device?

Panto works in any browser. To connect an Android device:

1. Plug in your device with USB
2. Enable developer mode and USB debugging mode on your android Device
3. Set the device to file transfer mode
4. In Panto, click Add Device
5. Choose USB device and select your device from the list

Once connected, the device is ready for test execution and automation. In case if Panto AI is opened more than one tabs, then it would throw an error message and additional tab would need to be closed

What is Panto Execute?

Panto Execute lets you interact with your mobile app using plain natural language. You give instructions and the Agent performs them on the device in real time.

Example for an add to cart test on the Blinkit app:

• Open Blinkit App
  • “Agent opens Blinkit"
• Search for Milk
  • “Agent searches for Milk”
• Add Milk to cart
  • “Agent adds Milk to cart”
• Click on View Cart

Shorter instructions lead to faster, more accurate execution.

What are Test Flows?

A Test Flow is a saved execution of a workflow. When the Agent completes execution of a feature correctly, click Save and assign a name. Saved flows can be searched, rerun, automated, and executed across multiple devices.

Different Buttons on Individual Test Flows

• Status: Shows the current state of a test flow. A flow can be In Progress while it is running, Passed when all steps complete successfully, or Failed if an issue is detected during execution. This helps teams quickly understand the health of each flow at a glance.

• View Video: Available once a run is completed. Clicking this opens a short playback of the exact steps executed on the device, making it easy to review behavior, validate flows, or understand why a run failed without digging into logs.

• Run: Opens the execution panel where you can choose the device, operating system, and app version. This allows the same flow to be run across different environments, devices, OS and systems without any changes to the test definition.

• View Static Code: Displays the auto generated Appium or Maestro code for the flow. Teams can directly edit or extend this code if they want more control, custom logic, or deeper integration with existing test suites. User can also get a quick summary of steps executed or regenerate the whole script.

• Modify Flow: You can combine multiple flows to create a complete end to end regression suite. Each flow represents a specific user journey, and Panto lets you stitch these journeys together to reflect real world app usage.

  Using pre flows and post flows, you can define what should run before and after a core test flow. For example, a pre-flow can handle common setup steps like login or onboarding, while a post flow can validate cleanup actions such as logout or state reset.

  Once combined, the entire regression suite can be executed as a single test across any device and OS combination. This approach reduces duplication, keeps flows modular, and makes it easy to maintain large test suites while still testing the full user journey end to end.

• Delete: Removes a flow that is no longer required, keeping the dashboard clean and focused on active and relevant tests.

What are Test Runs?

A Test Run is generated every time a Test Flow is executed on the device farm. It represents a single execution instance of that flow on a specific device, OS version, and app build.

Each Test Run clearly shows its current state, such as In Progress while execution is ongoing, Completed when all steps finish successfully, or Failed if any step encounters an issue. This makes it easy to track what is running, what has finished, and what needs attention.

Opening a Test Run provides a detailed, execution level view. You can inspect runtime logs, crash logs, and system output to understand failures or performance issues. A recorded video of the run is available to visually review how the app behaved on the real device.

The step by step execution timeline shows exactly how each action was performed, along with the static Appium or Maestro code generated for that specific run.

Test Runs act as the single source of truth for debugging, validation, and auditing, helping teams quickly move from failure detection to root cause analysis without switching tools.

How do I add a new app?

Click on Apps in the left pane, upload your APK, and the app becomes available for testing.

How do I run a test on multiple devices?

Go to Test Runs, choose the app, select the devices and OS versions you need, and click Run.

What is Context?

Context lets you give the Agent specific details that influence how tests should run. Provide this in natural language and the Agent will consider it during execution.

Examples

“The debug app password is 11111111”
“The cursor in the payment flow moves right to left”
“The language change option is located in settings” "package name of wikipedia app is org.wikipedia”

This ensures the Agent behaves in a way that matches real user experience.

What is Generate Static Code?

The Agent produces deterministic Appium or Maestro scripts, meaning every action executed during the natural language run is converted into stable, repeatable automation code. The same flow will always generate the same script, which removes flaky behavior and creates a reliable baseline for regression testing.

What are Integrations?

Integrations for Slack, Jira, and Confluence are being developed. They will be available in this section once released.

What are different types of Modes in Prompting?

• Autonomous: Autonomous mode lets you describe an entire workflow in a single prompt. The system understands the app structure and independently builds and executes the full flow step by step. It reasons using multiple signals including XML, visible text, images, and screen coordinates to decide the best action at each step. This mode is best suited for complex or non linear tasks that require deeper understanding and decision making, such as multi step user journeys or flows with dynamic UI changes.

Example

“Open the shopping app, search for running shoes under ₹3,000, apply size and brand filters, select the second product, add it to cart, log in with test credentials, and proceed till the payment screen.”

• Guided: Guided mode executes one instruction at a time by primarily relying on the underlying XML structure of the app. It is optimized for speed and predictability, making it up to five times faster than Autonomous mode for straightforward tasks. When the system does not have enough information to safely execute an instruction, it automatically falls back to Autonomous mode to reason through the scenario and continue execution without breaking the flow.

Example

“Tap on Search”
“Enter ‘running shoes’ in the search bar”
“Tap on the first result”