Install any skill in seconds. Free to start, no credit card required.
Get Started Free →将任意网页转换为桌面应用,支持 macOS/Windows/Linux 三大平台。使用 Rust + Tauri 技术栈,生成的应用体积小(约 5MB)、性能高。支持自定义图标、窗口大小、快捷键等丰富配置。
.claude/skills/anbeime-web-to-app/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 58% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 4% | 0% |
curl、wget、file、tar首次使用前,需要安装 pake-cli 工具:
bashnpm install -g pake-cli # 或使用 pnpm(推荐) pnpm install -g pake-cli
scripts/install_pake.py 检查并安装 pake-cliscripts/build_app.py 执行打包--hide-title-bar、--multi-instance)--debug 参数查看详细日志url:网页 URL(必需)name:应用名称icon:图标路径width:窗口宽度height:窗口高度options:其他可选参数(字典格式)./)生成应用安装包.dmg 安装包(设置 PAKE_CREATE_APP=1 可生成 .app).msi 安装包.deb 或 .AppImage 包--debug 参数查看详细日志--ignore-certificate-errors功能说明:将 GitHub 网页打包成应用 执行方式:脚本调用
pythonfrom scripts.build_app import build_app result = build_app( url="https://github.com", name="GitHub" ) # 输出:GitHub.dmg / GitHub_x64.msi / GitHub_x86_64.deb
功能说明:自定义窗口大小和图标 执行方式:脚本调用
pythonresult = build_app( url="https://chat.openai.com", name="ChatGPT", icon="https://example.com/icon.png", width=1400, height=900, options={ "hide-title-bar": True, "always-on-top": False } )
功能说明:允许同时打开多个应用窗口 执行方式:脚本调用
pythonresult = build_app( url="https://example.com", name="MyApp", options={ "multi-instance": True, "activation-shortcut": "CmdOrControl+Shift+P" } )
生成的应用内置以下快捷键:
| 操作 | macOS | Windows/Linux | |------|-------|---------------| | 刷新页面 | ⌘ + R | Ctrl + R | | 隐藏窗口 | ⌘ + W | Ctrl + W | | 放大/缩小 | ⌘ + +/- | Ctrl + +/- | | 重置缩放 | ⌘ + 0 | Ctrl + 0 | | 复制 URL | ⌘ + L | Ctrl + L | | 返回首页 | ⌘ + Shift + H | Ctrl + Shift + H | | 开发者工具 | ⌘ + Option + I | Ctrl + Shift + I(仅调试模式) | | 清除缓存重启 | ⌘ + Shift + Delete | Ctrl + Shift + Delete |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 16,067 | 11,547 | -28% | 1 | 1 | 0% | 3,000 | 3,670 | +22% | 0 | 0 | — |
case-01 | fail→pass | 9,171 | 5,009 | -45% | 1 | 1 | 0% | 1,952 | 2,559 | +31% | 0 | 0 | — |
case-02 | fail→fail | 20,440 | 2,322 | -89% | 1 | 1 | 0% | 3,102 | 1,801 | -42% | 0 | 0 | — |
case-03 | fail→pass | 8,988 | 5,218 | -42% | 1 | 1 | 0% | 1,747 | 2,353 | +35% | 0 | 0 | — |
case-04 | pass→pass | 13,308 | 8,805 | -34% | 1 | 1 | 0% | 2,392 | 3,206 | +34% | 0 | 0 | — |
case-06 | pass→pass | 11,026 | 10,035 | -9% | 1 | 1 | 0% | 2,142 | 2,954 | +38% | 0 | 0 | — |
case-07 | fail→pass | 12,490 | 6,109 | -51% | 1 | 1 | 0% | 1,709 | 2,705 | +58% | 0 | 0 | — |
case-08 | fail→pass | 15,432 | 2,800 | -82% | 1 | 1 | 0% | 2,200 | 1,985 | -10% | 0 | 0 | — |
case-09 | pass→pass | 9,219 | 5,347 | -42% | 1 | 1 | 0% | 2,000 | 2,493 | +25% | 0 | 0 | — |
case-10 | fail→pass | 11,249 | 5,822 | -48% | 1 | 1 | 0% | 2,532 | 2,635 | +4% | 0 | 0 | — |
case-11 | pass→pass | 8,327 | 2,922 | -65% | 1 | 1 | 0% | 1,638 | 1,946 | +19% | 0 | 0 | — |
case-12 | fail→fail | 51,596 | 4,416 | -91% | 1 | 1 | 0% | 3,118 | 1,693 | -46% | 0 | 0 | — |
case-13 | fail→pass | 3,724 | 1,511 | -59% | 1 | 1 | 0% | 631 | 1,615 | +156% | 0 | 0 | — |
case-14 | fail→pass | 7,319 | 1,768 | -76% | 1 | 1 | 0% | 1,172 | 1,662 | +42% | 0 | 0 | — |
case-15 | fail→pass | 10,975 | 2,539 | -77% | 1 | 1 | 0% | 1,614 | 1,885 | +17% | 0 | 0 | — |
case-16 | fail→pass | 7,519 | 1,770 | -76% | 1 | 1 | 0% | 1,323 | 1,690 | +28% | 0 | 0 | — |
case-17 | pass→pass | 3,259 | 2,300 | -29% | 1 | 1 | 0% | 550 | 1,699 | +209% | 0 | 0 | — |
case-18 | fail→pass | 7,362 | 1,431 | -81% | 1 | 1 | 0% | 1,282 | 1,600 | +25% | 0 | 0 | — |
case-19 | fail→pass | 4,536 | 1,973 | -57% | 1 | 1 | 0% | 704 | 1,762 | +150% | 0 | 0 | — |
case-20 | fail→pass | 4,898 | 2,007 | -59% | 1 | 1 | 0% | 961 | 1,752 | +82% | 0 | 0 | — |
case-21 | pass→pass | 10,944 | 7,836 | -28% | 1 | 1 | 0% | 2,107 | 2,955 | +40% | 0 | 0 | — |
case-22 | fail→pass | 14,525 | 1,605 | -89% | 1 | 1 | 0% | 2,751 | 1,690 | -39% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +59 percentage points is the difference between those two pass rates over the 21 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.