flutter-skill 0.8.3 → 0.8.5

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -1,6 +1,6 @@
1
1
  <p align="center">
2
2
  <a href="https://github.com/ai-dashboad/flutter-skill">
3
- <img src="assets/demo-teaser.gif" alt="AI testing a TikTok clone across 8 platforms" width="640">
3
+ <img src="assets/demo-teaser.gif" alt="AI testing a TikTok clone across 10 platforms" width="640">
4
4
  </a>
5
5
  </p>
6
6
 
@@ -8,7 +8,7 @@
8
8
 
9
9
  <p align="center">
10
10
  <strong>Give any AI agent eyes and hands inside any running app.</strong><br>
11
- 8 platforms. Zero test code. One MCP server.
11
+ 10 platforms. Zero test code. One MCP server.
12
12
  </p>
13
13
 
14
14
  <p align="center">
@@ -22,11 +22,15 @@
22
22
  <p align="center">
23
23
  <a href="#30-second-demo">Demo</a> •
24
24
  <a href="#quick-start">Quick Start</a> •
25
- <a href="#8-platforms-one-tool">Platforms</a> •
25
+ <a href="#10-platforms-one-tool">Platforms</a> •
26
26
  <a href="#why-not-playwright--appium--detox">vs Others</a> •
27
27
  <a href="docs/USAGE_GUIDE.md">Docs</a>
28
28
  </p>
29
29
 
30
+ <p align="center"><b>🚀 Zero config. Zero test code. Just talk to your AI.</b></p>
31
+
32
+ <p align="center"><sub>If this saves you time, please consider <a href="https://github.com/ai-dashboad/flutter-skill/stargazers">starring the repo ⭐</a> — it helps others find it!</sub></p>
33
+
30
34
  ---
31
35
 
32
36
  ## 30-Second Demo
@@ -39,9 +43,12 @@ https://github.com/user-attachments/assets/d4617c73-043f-424c-9a9a-1a61d4c2d3c6
39
43
 
40
44
  ## Why This Exists
41
45
 
42
- Writing E2E tests is painful. Maintaining them is worse.
46
+ Writing E2E tests is painful. Maintaining them is worse. **flutter-skill** takes a different approach:
43
47
 
44
- **flutter-skill** connects your AI agent (Claude, Cursor, Windsurf, Copilot, OpenClaw) directly to your running app via [MCP](https://modelcontextprotocol.io/). The agent can see the screen, tap buttons, type text, scroll — like a human tester who never sleeps.
48
+ - 🔌 **Connects any AI agent** (Claude, Cursor, Windsurf, Copilot, OpenClaw) directly to your running app via [MCP](https://modelcontextprotocol.io/)
49
+ - 👀 **The agent sees your screen** — taps buttons, types text, scrolls, navigates — like a human tester who never sleeps
50
+ - ✅ **Zero test code** — no Page Objects, no XPath, no brittle selectors. Just plain English
51
+ - ⚡ **Zero config** — 2 lines of code, works on all 10 platforms
45
52
 
46
53
  ```
47
54
  You: "Test the checkout flow with an empty cart, then add 3 items and complete purchase"
@@ -90,7 +97,7 @@ void main() {
90
97
 
91
98
  > *"Launch my app, explore every screen, and report any bugs"*
92
99
 
93
- That's it. **Zero configuration. Zero test code.**
100
+ That's it. **Zero configuration. Zero test code. Works in under 60 seconds.**
94
101
 
95
102
  <details>
96
103
  <summary>📦 More install methods (Homebrew, Scoop, Docker, IDE, Agent Skill)</summary>
@@ -111,9 +118,9 @@ That's it. **Zero configuration. Zero test code.**
111
118
 
112
119
  ---
113
120
 
114
- ## 8 Platforms, One Tool
121
+ ## 10 Platforms, One Tool
115
122
 
116
- Most testing tools work on 1-2 platforms. flutter-skill works on **8**.
123
+ Most testing tools work on 1-2 platforms. flutter-skill works on **10**.
117
124
 
118
125
  | Platform | SDK | Test Score |
119
126
  |----------|-----|:----------:|
@@ -125,11 +132,33 @@ Most testing tools work on 1-2 platforms. flutter-skill works on **8**.
125
132
  | **KMP Desktop** | [`sdks/kmp`](sdks/kmp/) | ✅ 75/75 |
126
133
  | **.NET MAUI** | [`sdks/dotnet-maui`](sdks/dotnet-maui/) | ✅ 75/75 |
127
134
  | **iOS** (Swift/UIKit) | [`sdks/ios`](sdks/ios/) | ✅ 19/19 |
135
+ | **Web** (any website) | [`sdks/web`](sdks/web/) | ✅ |
136
+ | **Web CDP** (zero-config) | No SDK needed | ✅ 141/156 |
128
137
 
129
138
  **Total: 656/664 tests passing (98.8%)** — each platform tested against a complex social media app with 50+ elements.
130
139
 
131
140
  ---
132
141
 
142
+ ## ⚡ Performance
143
+
144
+ Real benchmarks from automated test runs against a complex social media app:
145
+
146
+ | Operation | Web (CDP) | Electron | Android |
147
+ |-----------|:---------:|:--------:|:-------:|
148
+ | `connect` | 93 ms | 55 ms | 103 ms |
149
+ | `tap` | **1 ms** | **1 ms** | **2 ms** |
150
+ | `enter_text` | **1 ms** | **1 ms** | **2 ms** |
151
+ | `inspect` | 3 ms | 12 ms | 10 ms |
152
+ | `snapshot` | **2 ms** | **8 ms** | **29 ms** |
153
+ | `screenshot` | **31 ms** | **80 ms** | **88 ms** |
154
+ | `eval` | **1 ms** | — | — |
155
+
156
+ **Token efficiency:** `snapshot()` returns a structured element tree instead of an image — **87–99% fewer tokens** than sending screenshots to your AI agent.
157
+
158
+ **How fast is that?** A `tap` takes 1–2 ms end-to-end. Browser automation tools like Playwright and Selenium typically take 50–100 ms for the same operation. That's 50–100× faster, because flutter-skill talks directly to the app runtime instead of going through WebDriver or CDP indirection.
159
+
160
+ ---
161
+
133
162
  ## Why Not Playwright / Appium / Detox?
134
163
 
135
164
  | | flutter-skill | Playwright | Appium | Detox |
@@ -137,7 +166,7 @@ Most testing tools work on 1-2 platforms. flutter-skill works on **8**.
137
166
  | **Setup time** | 30 sec | Minutes | Hours | Hours |
138
167
  | **Test code needed** | ❌ None | ✅ Yes | ✅ Yes | ✅ Yes |
139
168
  | **AI-native (MCP)** | ✅ | ❌ | ❌ | ❌ |
140
- | **Platforms** | 8 | 3 (web) | Mobile | React Native |
169
+ | **Platforms** | 10 | 3 (web) | Mobile | React Native |
141
170
  | **Natural language** | ✅ | ❌ | ❌ | ❌ |
142
171
  | **Maintenance** | Zero | High | High | Medium |
143
172
  | **Flutter support** | ✅ Native | Partial | Partial | ❌ |
@@ -193,7 +222,7 @@ Most testing tools work on 1-2 platforms. flutter-skill works on **8**.
193
222
  </table>
194
223
 
195
224
  <details>
196
- <summary><strong>40+ tools — full reference</strong></summary>
225
+ <summary><strong>160+ tools — full reference</strong></summary>
197
226
 
198
227
  **Launch & Connect:** `launch_app`, `scan_and_connect`, `hot_reload`, `hot_restart`, `list_sessions`, `switch_session`, `close_session`
199
228
 
package/dart/pubspec.yaml CHANGED
@@ -1,6 +1,6 @@
1
1
  name: flutter_skill_npm
2
2
  description: Give your AI Agent eyes and hands inside your Flutter app.
3
- version: 0.2.0
3
+ version: 0.8.5
4
4
  publish_to: 'none'
5
5
  homepage: https://github.com/ai-dashboad/flutter-skill
6
6
  repository: https://github.com/ai-dashboad/flutter-skill
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "flutter-skill",
3
3
  "mcpName": "io.github.ai-dashboad/flutter-skill",
4
- "version": "0.8.3",
4
+ "version": "0.8.5",
5
5
  "description": "MCP Server for app automation - Give your AI Agent eyes and hands inside any app (Flutter, React, Web, Native)",
6
6
  "main": "index.js",
7
7
  "bin": {