Android UI Journey Testing
SkillDev toolsThis skill runs XML-specified user journeys on Android devices or emulators. Working as an android claude skill, it parses a test file, then executes each step in order using ADB commands for taps, swipes, and text input, with screenshots or UI dumps used for assertions.
Use Android UI Journey Testing in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add Android UI Journey Testing and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use the Android UI Journey Testing skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; ahel provides instructions and does not run this skill.
No other account needed.
Have an Android device or emulator connected and reachable through ADB.
What your AI can do with it
- Parses XML journey files where the <journey> root defines the test name and <actions>
- Performs taps at element coordinates via adb shell input tap
- Swipes and scrolls between coordinates with a duration via adb shell input swipe
- Types text into the active field via adb shell input text
- Verifies state assertions, using screenshots or UI dumps as evidence
- Writes a standardized JSON outcome report with pass/fail per step and skipped steps
Getting started
- Have an Android device or emulator connected and reachable through ADB.
- Write a journey XML file with a <journey> root, a name, and an <actions> block listing steps in order.
- Write interactive steps as actions (taps, swipes, text input) and assertion steps beginning with verify, check, or ensure.
- Ask your agent to run the skill against the XML file; it executes steps sequentially and stops at the first failure.
- Read the JSON report it outputs to see which steps passed, failed, or were skipped.
What this skill tells your AI
The instructions your AI receives, as published by sickn33/agentic-awesome-skills in skills/android-ui-journey-testing/SKILL.md and read by ahel’s review.
Overview
This skill outlines the standard workflow for running XML-specified User Journey tests on Android applications. A "journey" is a sequenced set of user actions and state assertions designed to verify end-to-end functionality. The journey XML acts as the source of truth for the app's behavior. The executor proceeds sequentially, performing UI interactions, checking state assertions, and writing a standardized JSON outcome report.
When to Use
- Use when evaluating an Android application's UI behavior against an XML test specification.
- Use when automating multi-step user flows (e.g., login, checkout, navigation) and verifying expectations.
- Use when running verification tests and generating standardized JSON test reports.
- Use when debugging application flows on physical devices or emulators using ADB commands.
How It Works
graph TD
A[Parse Journey XML] --> B[Execute Action / Interaction]
B --> C{Success?}
C -- Yes --> D[Verify State Expectation / Assertion]
C -- No/Crash --> G[Mark FAILED & Exit]
D -- Passed --> E{More Steps?}
D -- Failed --> G
E -- Yes --> B
E -- No --> F[Output JSON Summary]
Step 1: Parse the Journey Specification
Read and parse the XML test suite structure. The root node <journey> defines the test case name, and the <actions> block contains the sequence of test steps.
<journey name="Search and Cart Flow">
<description>Verify that searching for an item and adding it to the cart succeeds.</description>
<actions>
<action>Search for soda</action>
<action>Tap the first search result</action>
<action>Verify that the product detail screen is shown</action>
</actions>
</journey>
Step 2: Sequential Step Evaluation
Process each <action> element in the exact order specified. Test steps are classified into two groups:
A. Interactive Actions (Taps, Swipes, Text Input)
Perform the physical UI interaction using ADB.
- Tapping: Tap the center of the target element's bounds:
adb shell input tap <x> <y> - Swiping/Scrolling: Swipe from coordinate to coordinate with a duration:
adb shell input swipe <x1> <y1> <x2> <y2> <duration_ms> - Text Typing: Type text into the active input field:
adb shell input text "<string>"
If the element is missing or the action cannot be performed, the action and the journey fail.
B. State Assertions (Expectation Verification)
Steps beginning with "verify", "check", or "ensure" represent state assertions.
- Inspect the current screen (using screenshots or
uiautomator dump) without interacting or scrolling. - Confirm all sub-assertions are met. For example, "Verify that the app is on the Home screen and the logo is visible" fails if the home screen is not displayed or the logo is missing.
Step 3: Handle Failures and Crashes
If the application crashes, exits, freezes, or fails an assertion:
- Immediately stop journey execution.
- Mark the failed step as
FAILED. - Mark any subsequent steps as
SKIPPED. - Document the exact reason for failure.
Step 4: Generate JSON Report
Format the execution results into a standardized JSON schema and write it to the output log.
Examples
Example 1: Full Journey XML Specification
<journey name="Login and Profile Edit">
<description>Logs into the app, navigates to settings, and changes user profile information.</description>
<actions>
<action>Verify that the username input field is visible</action>
<action>Tap the username input field</action>
<action>Type "testuser" into the input</action>
<action>Tap the password input field</action>
<action>Type a redacted test password into the input</action>
<action>Tap the "Login" button</action>
<action>Verify that the Home dashboard is visible and user profile photo is shown</action>
</actions>
</journey>
Example 2: Standard JSON Outcome Report
{
"journey": "Login and Profile Edit",
"results": [
{
"action": "Verify that the username input field is visible",
"status": "PASSED",
"commands": [],
"comment": "Username input detected at bounds [100,200][980,300] via UI dump."
},
{
"action": "Tap the username input field",
"status": "PASSED",
"commands": [
"adb shell input tap 540 250"
],
"comment": "Tapped center coordinates of username input."
},
{
"action": "Type \"testuser\" into the input",
"status": "PASSED",
"commands": [
"adb shell input text \"testuser\""
],
"comment": "Username typed successfully."
},
{
"action": "Tap the password input field",
"status": "PASSED",
"commands": [
"adb shell input tap 540 370"
],
"comment": "Tapped center of password input."
},
{
"action": "Type a redacted test password into the input",
"status": "PASSED",
"commands": [
"adb shell input text \"[REDACTED_PASSWORD]\""
],
"comment": "Password typed successfully. The actual input value was not stored in the report."
},
{
"action": "Tap the \"Login\" button",
"status": "PASSED",
"commands": [
"adb shell input tap 540 500"
],
"comment": "Login button clicked."
},
{
"action": "Verify that the Home dashboard is visible and user profile photo is shown",
"status": "FAILED",
"commands": [],
"comment": "Dashboard loaded but profile photo was missing from the UI header."
}
]
}
Best Practices
- ✅ Calculate Centers for Taps: When parsing element bounds like
[x1,y1][x2,y2], always compute the middle coordinate: $$x_{center} = \frac{x_1 + x_2}{2}, \quad y_{center} = \frac{y_1 + y_2}{2}$$ - ✅ Include Sleep Buffers: Always add a short delay (e.g., 1-2 seconds) after interactive actions (like button taps) to let layouts and transitions render before executing assertions.
- ✅ Fail Fast: Stop the test immediately upon encountering the first failure. Continuing after a failure leads to invalid results.
- ✅ Log Precise Commands Safely: Include non-sensitive raw commands (such as
adb shell input tap) in the JSON output list for diagnostics. Redact text entered into password, OTP, token, payment, or personal-data fields; never persist the literal secret in reports, CI logs, or shared artifacts.
Limitations
- The parser only evaluates the static screen hierarchy (e.g.
uiautomator dump). Elements that require scrolling are marked as not visible unless a scrolling action is explicitly performed. - Non-standard UI components (like custom OpenGL canvas views) cannot be read via standard accessibility trees and may require screenshot analysis or hardcoded click maps.
- Key events and text typing via ADB do not trigger standard soft keyboard events on all emulator images, which can lead to input validation issues.
Related Skills
@android-cli- General CLI tool syntax, package install, and device queries.@android_ui_verification- Direct ADB script templates for general UI checks.
Signals
- GitHub stars
- 47k
- Forks
- 7k
- Last commit
- Sep 2026
Others that do the same job
Questions
- Can Claude use skills on mobile?
- This skill targets Android devices and emulators, driving them over ADB from an agent session. Whether skills run inside a mobile app depends on that app; the item lists claude, cursor, gemini, and antigravity as supported tools.
- Can I use Claude AI on Android?
- This item does not describe Claude's own mobile availability. It is a testing skill that automates UI journeys on Android devices or emulators through ADB commands.
Advanced
- Item type
- skill
- Key
android-ui-journey-testing-sickn33- Source
- github.com/sickn33/agentic-awesome-skills
github.com/sickn33/agentic-awesome-skills
Related picks
Skill · sickn33
Does the same job in other wordshandsontable-playwright-e2e
Skill · handsontable
The pick for End-to-end testingmstar-e2e
Skill · btspoony
The pick for End-to-end testingteach
Skill · mattpocock
More in Dev toolsimplement
Skill · mattpocock
More in Dev toolsponytail
Skill · dietrichgebert
More in Dev tools