원클릭으로 Manus에서 모든 스킬 실행

시작하기

maestro

스타8

포크1

업데이트2026년 2월 9일 19:11

Create Maestro UI test flows for user stories and features

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

michele-prandina

michele-prandina/ClaudE2E

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

Maestro UI Testing Skill

Create Maestro test flows for user stories and features.

Steps

Identify the test scope:
- Single story → create {test_dir}/S{XX}-{feature}.yaml
- Reusable flow (login, navigation) → create {test_dir}/flows/{name}.yaml
- Full feature suite → create multiple flows + index file
Determine the app ID:
- Check your project's app config for the bundle identifier
- Default: {your.app.bundleIdentifier}

Write the flow using Maestro YAML syntax:

appId: {your.app.bundleIdentifier}
---
# Test: {Story title or feature description}

- launchApp
- assertVisible: "Expected screen element"
- tapOn: "Button text"
- inputText: "test input"
- assertVisible: "Success message"

Use appropriate selectors (in order of preference):
- text: "Visible text" — most readable, use for static content
- id: "accessibility_id" — use for dynamic content or icons
- point: 50%, 50% — last resort for complex layouts
Include standard test patterns:
- Happy path (main user flow)
- Error states (invalid input, network errors)
- Edge cases (empty states, long text)

Maestro Command Reference

Command	Syntax	Use Case
`launchApp`	`- launchApp`	Start the app
`tapOn`	`- tapOn: "Text"`	Tap button/element
`inputText`	`- inputText: "value"`	Type in focused field
`assertVisible`	`- assertVisible: "Text"`	Verify element exists
`assertNotVisible`	`- assertNotVisible: "Text"`	Verify element hidden
`scroll`	`- scroll`	Scroll down
`swipe`	`- swipe: {direction: "LEFT"}`	Swipe gesture
`back`	`- back`	Android back button
`hideKeyboard`	`- hideKeyboard`	Dismiss keyboard
`takeScreenshot`	`- takeScreenshot: "name"`	Capture screen
`runFlow`	`- runFlow: "flows/login.yaml"`	Reuse another flow
`waitForAnimationToEnd`	`- waitForAnimationToEnd`	Wait for animations

Selector Options

# By text (preferred)
- tapOn: "Submit"

# By accessibility ID
- tapOn:
    id: "submit_button"

# By text with index (when multiple matches)
- tapOn:
    text: "Item"
    index: 0

# By position relative to other elements
- tapOn:
    below: "Header"
    text: "Button"

# By screen coordinates (last resort)
- tapOn:
    point: 50%, 90%

Usage

/maestro S42              # Create test flow for story S42
/maestro login            # Create reusable login flow
/maestro "onboarding"     # Create test for onboarding feature

Output

Flow file created in the project's test directory
Includes assertions for acceptance criteria
Ready to run with maestro test {path-to-flow}.yaml

Running Tests

After creating flows, run them:

# Single flow
maestro test {test_dir}/S42-feature.yaml

# All flows
maestro test {test_dir}/

# With screenshots on failure
maestro test --debug-output ./screenshots {test_dir}/S42-feature.yaml

Best Practices

One flow per user journey — not one flow per screen
Use text selectors — they read like user instructions
Add waits after navigation — waitForAnimationToEnd or explicit delays
Reuse login flows — runFlow: "flows/login.yaml"
Test error states — not just happy paths
Keep flows short — 10-20 steps max, split if longer

이 저장소의 다른 Skills

같은 저장소

agent-stories

michele-prandina/ClaudE2E

Write user stories optimized for AI coding agents. Use this skill when writing backlog stories that will be implemented by Developer or Frontend Developer agents. Produces structured, XML-tagged specifications with explicit context, states matrix, file paths, and escalation criteria. Based on research from SWE-bench, MetaGPT, and practitioner evidence from Claude Code, Cursor, and Devin communities.

2026-02-098

deep-research

michele-prandina/ClaudE2E

Multi-phase deep research with parallel retrieval agents. Use when the user asks for in-depth research, analysis, comparison, market intelligence, or any query that requires systematic evidence gathering from multiple sources with citations and confidence scoring. Triggers on: "deep research", "research this", "investigate", "comprehensive analysis", "compare X vs Y in depth", "what does the evidence say about", or any complex question needing 5+ sources.

2026-02-098

Maestro UI Testing Skill

Create Maestro test flows for user stories and features.

Steps

Identify the test scope:

Single story → create {test_dir}/S{XX}-{feature}.yaml
Reusable flow (login, navigation) → create {test_dir}/flows/{name}.yaml
Full feature suite → create multiple flows + index file

Determine the app ID:

Check your project's app config for the bundle identifier
Default: {your.app.bundleIdentifier}

Write the flow using Maestro YAML syntax:

appId: {your.app.bundleIdentifier}
---
# Test: {Story title or feature description}

- launchApp
- assertVisible: "Expected screen element"
- tapOn: "Button text"
- inputText: "test input"
- assertVisible: "Success message"

Use appropriate selectors (in order of preference):

text: "Visible text" — most readable, use for static content
id: "accessibility_id" — use for dynamic content or icons
point: 50%, 50% — last resort for complex layouts

Include standard test patterns:

Happy path (main user flow)
Error states (invalid input, network errors)
Edge cases (empty states, long text)

Maestro Command Reference

Command

Syntax

Use Case

launchApp

- launchApp

Start the app

tapOn

- tapOn: "Text"

Tap button/element

inputText

- inputText: "value"

Type in focused field

assertVisible

- assertVisible: "Text"

Verify element exists

assertNotVisible

- assertNotVisible: "Text"

Verify element hidden

scroll

- scroll

Scroll down

swipe

- swipe: {direction: "LEFT"}

Swipe gesture

back

- back

Android back button

hideKeyboard

- hideKeyboard

Dismiss keyboard

takeScreenshot

- takeScreenshot: "name"

Capture screen

runFlow

- runFlow: "flows/login.yaml"

Reuse another flow

waitForAnimationToEnd

- waitForAnimationToEnd

Wait for animations

Selector Options

# By text (preferred) - tapOn: "Submit" # By accessibility ID - tapOn: id: "submit_button" # By text with index (when multiple matches) - tapOn: text: "Item" index: 0 # By position relative to other elements - tapOn: below: "Header" text: "Button" # By screen coordinates (last resort) - tapOn: point: 50%, 90%

Usage

/maestro S42 # Create test flow for story S42 /maestro login # Create reusable login flow /maestro "onboarding" # Create test for onboarding feature

Output

Flow file created in the project's test directory

Includes assertions for acceptance criteria

Ready to run with maestro test {path-to-flow}.yaml

Running Tests

After creating flows, run them:

# Single flow maestro test {test_dir}/S42-feature.yaml # All flows maestro test {test_dir}/ # With screenshots on failure maestro test --debug-output ./screenshots {test_dir}/S42-feature.yaml

Best Practices

One flow per user journey — not one flow per screen

Use text selectors — they read like user instructions

Add waits after navigation — waitForAnimationToEnd or explicit delays

Reuse login flows — runFlow: "flows/login.yaml"

Test error states — not just happy paths

Keep flows short — 10-20 steps max, split if longer

name	maestro
description	Create Maestro UI test flows for user stories and features
user_invocable	true