e2e-testing

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

AI E2E Testing — 8 Platforms, Zero Test Code

AI端到端测试——支持8大平台，无需编写测试代码

Give your AI agent eyes and hands inside any running app.

flutter-skill is an MCP server that connects AI agents to running apps. The agent can see screenshots, tap elements, enter text, scroll, navigate, inspect UI trees, and verify state — all through natural language.

为你的AI Agent赋予在任意运行中应用内的「眼」和「手」。

flutter-skill是一款MCP服务器，可将AI Agent与运行中的应用连接。Agent能够通过自然语言完成截图、点击元素、输入文本、滚动、导航、检查UI树以及验证状态等操作。

Supported Platforms

支持的平台

Platform	Setup
Flutter (iOS/Android/Web)	`flutter pub add flutter_skill`
React Native	`npm install flutter-skill-react-native`
Electron	`npm install flutter-skill-electron`
iOS (Swift/UIKit)	SPM: `FlutterSkillSDK`
Android (Kotlin)	Gradle: `flutter-skill-android`
Tauri (Rust)	`cargo add flutter-skill-tauri`
KMP Desktop	Gradle dependency
.NET MAUI	NuGet package

Test scorecard: 562/567 (99.1%) across all 8 platforms.

平台	配置方式
Flutter (iOS/Android/Web)	`flutter pub add flutter_skill`
React Native	`npm install flutter-skill-react-native`
Electron	`npm install flutter-skill-electron`
iOS (Swift/UIKit)	SPM: `FlutterSkillSDK`
Android (Kotlin)	Gradle: `flutter-skill-android`
Tauri (Rust)	`cargo add flutter-skill-tauri`
KMP Desktop	Gradle依赖
.NET MAUI	NuGet包

测试评分：全8大平台共567项测试，通过562项（通过率99.1%）。

Install

安装

bash

undefined

bash

undefined

npm (recommended)

npm（推荐）

npm install -g flutter-skill

Homebrew

brew install ai-dashboad/flutter-skill/flutter-skill

Or download binary from GitHub Releases

或从GitHub Releases下载二进制文件

undefined

undefined

MCP Configuration

MCP配置

Add to your AI agent's MCP config (Claude Desktop, Cursor, Windsurf, OpenClaw, etc.):

json

{
  "mcpServers": {
    "flutter-skill": {
      "command": "flutter-skill",
      "args": ["server"]
    }
  }
}

添加到你的AI Agent的MCP配置中（Claude Desktop、Cursor、Windsurf、OpenClaw等）：

json

{
  "mcpServers": {
    "flutter-skill": {
      "command": "flutter-skill",
      "args": ["server"]
    }
  }
}

OpenClaw

If using OpenClaw, add to your gateway config under

mcp.servers

yaml

mcp:
  servers:
    flutter-skill:
      command: flutter-skill
      args: ["server"]

如果使用OpenClaw，在网关配置的

mcp.servers

下添加：

yaml

mcp:
  servers:
    flutter-skill:
      command: flutter-skill
      args: ["server"]

Quick Start

快速开始

1. Initialize your app (one-time)

1. 初始化你的应用（仅需一次）

bash

cd /path/to/your/app
flutter-skill init

Auto-detects project type and patches your app with the testing bridge.

bash

cd /path/to/your/app
flutter-skill init

自动检测项目类型，并为你的应用添加测试桥接补丁。

2. Launch and connect

2. 启动并连接

bash

flutter-skill launch .

bash

flutter-skill launch .

3. Test with natural language

3. 使用自然语言测试

Tell the agent what to test:

"Test the login flow — enter admin@test.com and password123, tap Login, verify Dashboard appears"

The agent will automatically:

```
screenshot()
```
→ see the current screen
```
inspect_interactive()
```
→ discover all tappable/typeable elements with semantic refs
```
tap(ref: "button:Login")
```
→ tap using stable semantic reference

enter_text(ref: "input:Email", text: "admin@test.com")

→ type into field

```
wait_for_element(key: "Dashboard")
```
→ verify navigation
```
screenshot()
```
→ confirm final state

告诉Agent你要测试的内容：

"测试登录流程——输入admin@test.com和password123，点击登录按钮，验证仪表盘页面是否出现"

Agent会自动执行以下操作：

```
screenshot()
```
→ 捕获当前屏幕
```
inspect_interactive()
```
→ 发现所有可点击/可输入元素及其语义引用
```
tap(ref: "button:Login")
```
→ 使用稳定的语义引用点击按钮

enter_text(ref: "input:Email", text: "admin@test.com")

→ 在输入框中输入文本

```
wait_for_element(key: "Dashboard")
```
→ 验证页面导航
```
screenshot()
```
→ 确认最终状态

Available MCP Tools

可用的MCP工具

Core Actions

核心操作

Tool	Description
`screenshot`	Capture current screen as image
`tap`	Tap element by key, text, ref, or coordinates
`enter_text`	Type text into a field
`scroll`	Scroll up/down/left/right
`swipe`	Swipe gesture between points
`long_press`	Long press an element
`drag`	Drag from point A to B
`go_back`	Navigate back
`press_key`	Send keyboard key events

工具	描述
`screenshot`	捕获当前屏幕为图片
`tap`	通过键、文本、引用或坐标点击元素
`enter_text`	在输入框中输入文本
`scroll`	向上/向下/向左/向右滚动
`swipe`	在两点间执行滑动手势
`long_press`	长按元素
`drag`	从A点拖动到B点
`go_back`	返回上一页
`press_key`	发送键盘按键事件

Inspection (v0.8.0+)

检查功能（v0.8.0+）

Tool	Description
`inspect_interactive`	NEW — Get all interactive elements with semantic ref IDs
`get_elements`	List all elements on screen
`find_element`	Find element by key or text
`wait_for_element`	Wait for element to appear (with timeout)
`get_element_properties`	Get detailed properties of an element

工具	描述
`inspect_interactive`	新增 — 获取所有带语义引用ID的可交互元素
`get_elements`	列出屏幕上的所有元素
`find_element`	通过键或文本查找元素
`wait_for_element`	等待元素出现（带超时机制）
`get_element_properties`	获取元素的详细属性

Text Manipulation

文本操作

Tool	Description
`set_text`	Replace text in a field
`clear_text`	Clear a text field
`get_text`	Read text content

工具	描述
`set_text`	替换输入框中的文本
`clear_text`	清空输入框
`get_text`	读取文本内容

App Control

应用控制

Tool	Description
`get_logs`	Read app logs
`clear_logs`	Clear log buffer

工具	描述
`get_logs`	读取应用日志
`clear_logs`	清空日志缓冲区

Semantic Refs (v0.8.0)

语义引用（v0.8.0）

inspect_interactive

returns elements with stable semantic reference IDs:

button:Login          → Login button
input:Email           → Email text field
toggle:Dark Mode      → Dark mode switch
button:Submit[1]      → Second Submit button (disambiguated)

Format:

{role}:{content}[{index}]

7 roles:

button

input

toggle

slider

select

link

item

Use refs for reliable element targeting that survives UI changes:

tap(ref: "button:Login")
enter_text(ref: "input:Email", text: "test@example.com")

inspect_interactive

会返回带有稳定语义引用ID的元素：

button:Login          → 登录按钮
input:Email           → 邮箱输入框
toggle:Dark Mode      → 暗黑模式开关
button:Submit[1]      → 第二个提交按钮（已区分）

格式：

{角色}:{内容}[{索引}]

7种角色：

button

、

input

、

toggle

、

slider

、

select

、

link

、

item

使用引用进行可靠的元素定位，即使UI发生变化也不受影响：

tap(ref: "button:Login")
enter_text(ref: "input:Email", text: "test@example.com")

Testing Workflow

测试工作流

Basic Flow

基础流程

screenshot() → inspect_interactive() → tap/enter_text → screenshot() → verify

screenshot() → inspect_interactive() → tap/enter_text → screenshot() → 验证

Comprehensive Testing

全面测试

"Explore every screen of this app. Test all buttons, forms, navigation, and edge cases. Report any bugs you find."

The agent will systematically:

Navigate every screen via tab bars, menus, links
Interact with every interactive element
Test form validation (empty, invalid, valid inputs)
Test edge cases (long text, special characters, emoji)
Verify navigation flows (forward, back, deep links)
Take screenshots at each step for verification

"探索应用的每一个页面。测试所有按钮、表单、导航和边缘情况。报告发现的任何Bug。"

Agent会系统性地执行以下操作：

通过标签栏、菜单、链接导航到每一个页面
与所有可交互元素进行交互
测试表单验证（空值、无效值、有效值输入）
测试边缘情况（长文本、特殊字符、表情符号）
验证导航流程（前进、后退、深度链接）
在每一步截图以便验证

Example Prompts

示例提示词

Quick smoke test:

"Tap every tab and screenshot each page"

Form testing:

"Fill the registration form with edge case data — emoji name, very long email, short password — and verify error messages"

Navigation:

"Test the complete user journey: sign up → create post → like → comment → delete → sign out"

Accessibility:

"Check every screen for missing labels, small tap targets, and contrast issues"

快速冒烟测试：

"点击每个标签页并截图每个页面"

表单测试：

"用边缘情况数据填写注册表单——表情符号名称、超长邮箱、短密码——并验证错误提示"

导航测试：

"测试完整用户旅程：注册→创建帖子→点赞→评论→删除→登出"

无障碍测试：

"检查每个页面是否存在缺失标签、点击目标过小和对比度问题"

Tips

小贴士

Always start with
screenshot()
— see before you act
Use
inspect_interactive()
to discover elements — don't guess at selectors
Prefer
ref:
selectors — more stable than text or coordinates
wait_for_element()
after navigation — apps need time to transition
Screenshot after every action — verify the expected effect
Use
press_key
for keyboard shortcuts — test keyboard navigation

始终从
screenshot()
开始 — 先查看再操作
使用
inspect_interactive()
发现元素 — 不要猜测选择器
优先使用
ref:
选择器 — 比文本或坐标更稳定
导航后使用
wait_for_element()
— 应用需要时间完成过渡
每次操作后截图 — 验证预期效果
使用
press_key
测试键盘快捷键 — 测试键盘导航

e2e-testing

Original

Translation

AI E2E Testing — 8 Platforms, Zero Test Code

AI端到端测试——支持8大平台，无需编写测试代码

Supported Platforms

支持的平台

Install

安装

npm (recommended)

npm（推荐）

Homebrew

Homebrew

Or download binary from GitHub Releases

或从GitHub Releases下载二进制文件

MCP Configuration

MCP配置

OpenClaw

OpenClaw

Quick Start

快速开始

1. Initialize your app (one-time)

1. 初始化你的应用（仅需一次）

2. Launch and connect

2. 启动并连接

3. Test with natural language

3. 使用自然语言测试

Available MCP Tools

可用的MCP工具

Core Actions

核心操作

Inspection (v0.8.0+)

检查功能（v0.8.0+）

Text Manipulation

文本操作

App Control

应用控制

Semantic Refs (v0.8.0)

语义引用（v0.8.0）

Testing Workflow

测试工作流

Basic Flow

基础流程

Comprehensive Testing

全面测试

Example Prompts

示例提示词

Tips

小贴士

Links

相关链接