Huashu Mac Use

操控没有API的macOS原生app并留下可复现取证：截指定窗口（后台被遮挡也能截）、点击输入、探测某个app能不能自动化、内嵌Chromium的app走CDP零焦点操控、做产品实测与多模型跑批。浏览器里的事走huashu-chrome，翻旧截图走disk-keeper。

330 stars 52 forksSoftware Development

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add alchaincyf/huashu-mac-use --skill huashu-mac-use

Manual install

git clone --depth 1 https://github.com/alchaincyf/huashu-mac-use.git /tmp/huashu-mac-use
mkdir -p ~/.claude/skills
cp -r /tmp/huashu-mac-use/. ~/.claude/skills/huashu-mac-use

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# huashu-mac-use · 操控没有 API 的原生 app

你是 macOS 自动化工程师兼取证员。你的独特价值不是「能截图」，是**操控一个既没有 CLI 也没有 API 的 GUI-only 原生 app，并把每一步都留成可复现的证据**——桌面 AI 客户端这类东西，命令行进不去、网页工具够不着，只有你能一边操作一边把窗口拍下来。你交付的标准是：用户在另一个窗口打字时完全感觉不到你在干活，而你交出的每一张图都能说出它是怎么来的。

`$SKILL_DIR` 指本文件所在目录。脚本一律写全路径调用（agent 的 cwd 是用户项目目录）。首次用先 `bash $SKILL_DIR/scripts/build.sh` 编译内核（约 20 秒）。

## 一、第一步该不该是它

| 任务信号 | 走哪 |
|---|---|
| 操控 Mac 原生 app：点菜单、填输入框、读界面、切窗口；系统层动作（打开退出 app、系统设置） | 本 skill |
| 给某个 app 的窗口截图取证（配图、汇报、复现 bug） | 本 skill，窗口级截图别的 skill 都够不着 |
| 探测陌生 app 能不能自动化 | `probe.sh` |
| 多模型同题跑批，目标是**桌面客户端本身** | 本 skill（只关心模型能力 → multi-model 走 API） |
| 浏览器里的任何事 | huashu-chrome。Chrome/Safari/Edge/Arc 的窗口即使列出来也不接 |
| 验收自己写的本地 HTML | webapp-testing；别人 app 里产出的 deck 只能截窗口 → 本 skill，视口对齐设计尺寸 |
| 找已有截图、批量 OCR | disk-keeper。剪录屏 → huastudio-edit / huashu-jianji。压图 → image-compress |

## 二、心智模型：四层控制面，读写分离

| 层 | 手段 | 何时 |
|---|---|---|
| L0 结构接口 | `cdp.js`（内嵌 Chromium 的 app）、AppleScript 字典、URL scheme、本地端口 | **默认起点**。零焦点、跨 Space、多 agent 互不干扰 |
| L1 AX 树 | `mac ax` / `mac axset` | 探到可编辑控件且实写后状态指示器变化才算通 |
| L2 窗口坐标 | `mac see` → `mac op` / `mac clickin` | 前两层不可用时的主力，需借焦点 |
| L3 像素 | `mac shot` | 控制手段的最后一档；**验证手段的每一步** |

三条原则，压过一切效率考虑：

Read the full skill on GitHub

Details

Skill name: huashu-mac-use
Repository: alchaincyf/huashu-mac-use
License: MIT
Repo updated: 13 October 2026

FAQ

What does the Huashu Mac Use skill do?

How do I install the Huashu Mac Use skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add alchaincyf/huashu-mac-use --skill huashu-mac-use. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Huashu Mac Use popular?

It lives in alchaincyf/huashu-mac-use, which has 330 GitHub stars and 52 forks.

Related skills

Pixel Path

283

jonwheatley/pixel-path

Pixel Path turns any two places (optionally via waypoints) into a pixel-art fly-through of the real road. It routes the trip, bakes the real buildings, land cover, landmarks, hills and horizon from OpenStreetMap and elevation data, generates scenery art with RetroDiffusion at a chosen level of detail (sparse, moderate or detailed; estimated up front, quoted exactly for free, budget-capped; free placeholder art also works), builds a page where you zoom along the route and scroll to go faster or slower, and runs a self-check that screenshots the result. Use when someone asks for a "pixel path",

Software Development19

Dr Jskill

344

jdubois/dr-jskill

Creates Java + Spring Boot projects: Web applications, full-stack apps with Vue.js or Angular or React or vanilla JS, PostgreSQL, REST APIs, and Docker. Use when creating Spring Boot projects, setting up Java microservices, or building enterprise applications with the Spring Framework.

Software Development58

Vibex

389

Raja0sama/vibex

Diagrams and checkable docs from a codebase. Draws ERDs, C4 context/container/component views, API endpoint catalogues, lifecycle/state machines, sequence diagrams, and service-to-service call maps as a validated JSON spec plus a standalone HTML viewer, with Mermaid export. Imports OpenAPI, GraphQL SDL and Prisma; otherwise reads the code (NestJS, TypeORM, Express) and writes the spec. Use for: database diagram, ERD, data model, table relationships, C4 or architecture diagram, system context, API map, endpoint list, "show me the endpoints", status flow, state machine, allowed transitions, "wha

Software Development

Robot Framework Skill

386

LambdaTest/agent-skills

Generates Robot Framework tests in keyword-driven syntax with Python. Supports SeleniumLibrary, RequestsLibrary, and custom keywords. Use when user mentions "Robot Framework", "*** Test Cases ***", "SeleniumLibrary", ".robot file". Triggers on: "Robot Framework", "*** Test Cases ***", ".robot", "SeleniumLibrary", "keyword-driven test".

Software DevelopmentIn 1 list76

Wechat Use

210

leeguooooo/wechat-use

Use WeChat on macOS Apple Silicon or Windows x64: query sessions, contacts and history, listen for messages, and send authorized text. Select the host-specific backend first. Windows has a standalone experimental CLI/MCP package with no system Python requirement; macOS uses the managed WeChat 4.1.9 clone and supports HTTP Bridge/Wechaty.

Software Development136

Phpunit Skill

386

LambdaTest/agent-skills

Generates PHPUnit tests in PHP. Covers assertions, data providers, mocking, and test doubles. Use when user mentions "PHPUnit", "TestCase", "assertEquals", "PHP test". Triggers on: "PHPUnit", "TestCase PHP", "assertEquals PHP", "PHP unit test".

Software DevelopmentIn 1 list76

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.