Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add cass-2003/local-workflow-skill --skill autojs-automationgit clone --depth 1 https://github.com/cass-2003/local-workflow-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cass-2003/local-workflow-skill/autojs-automation)<a href="https://agentmods.dev/skills/cass-2003/local-workflow-skill/autojs-automation"><img src="https://agentmods.dev/badge/skills/cass-2003/local-workflow-skill/autojs-automation.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00112 | $0.07396 |
| Opus 5 | $0.00056 | $0.03698 |
| Sonnet 5 | $0.00022 | $0.01479 |
| Haiku 4.5 | $0.00011 | $0.00740 |
Grade A, and why
autojs-automation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 221 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AutoJS 自动化
AutoJS 自动化（autojs-automation，兼容 slug: ajs）负责本技能描述范围内的定位、执行、验证和交接边界；旧短 slug 仅作兼容 alias/URL 主键，不作为规范技能名。
定位：把 Auto.js / AutoX.js / AutoJS Pro 自动化从“脚本能跑”收敛成可复核排障闭环：版本/设备/权限/入口/证据 → 场景执行卡 → 高频坑/防遗漏 → 输出要求 → 约束 → 反例库 → 提交前自检 → 2024-2026 新坑速查 → 相邻技能边界。 铁律：合法自有设备、授权测试、无障碍辅助优先；无版本、设备、权限、控件树、截图或稳定性日志证据，不宣称已修复；严禁恶意刷量、绕风控、账号滥用、抢购作弊、规避平台规则、窃取数据或攻击第三方 App。
快速总则
单技能工程门禁
- 先判定是否合法：只允许自有设备、授权测试、无障碍辅助、企业内控巡检和设备运维；批量账号、刷量、抢购、绕风控、验证码/实名/支付绕过直接拒绝。
- 先锁定运行画像：设备、ROM、Android、目标 App、AutoJS 引擎、权限、分辨率、字体缩放、网络、登录态和页面入口未明确，不得宣称脚本可稳定运行。
- 先定义成功/失败判据：每个 step 都要有进入条件、动作、业务信号等待、超时、重试上限、失败证据、人工接管和可恢复状态。
- 先做幂等保护：涉及提交、删除、支付确认、发布、发消息、改资料等副作用动作时，必须有唯一键、二次确认、结果回读和不确定即停。
- 先做真机矩阵：至少覆盖目标最低 Android、主力厂商 ROM、目标分辨率/字体、深浅色、弱网、锁屏/前后台和权限弹窗；未测设备只列风险。
- 先做证据闭环：运行日志、截图/录屏、控件树摘要、选择器、bounds、Activity、runId、设备和异常堆栈必须能复盘。
- 先做合规边界：所谓反检测只能用于降低误触、误封、误报和测试噪声；不得提供绕过平台风控、验证码、实名、支付确认、账号限制、反作弊或安全检测的实现路径。
低级错禁止清单
- 禁止固定 sleep 代替 waitFor 业务信号。
- 禁止绝对坐标作为首选路径；坐标兜底必须有区域校验、比例换算和动作后确认。
- 禁止 findOne 无超时、无限滑动、无限重试、高频截图/OCR 或死循环保活。
- 禁止只测模拟器、云手机或单台旗舰机就说兼容真机矩阵。
- 禁止未检查无障碍/截图/悬浮窗/通知/前台服务/电量优化就改 selector。
- 禁止把 OCR、找图、旧截图或控件树空结果当确定事实。
- 禁止日志写入账号、手机号、token、cookie、身份证、未脱敏截图或敏感 OCR 文本。
- 禁止脚本失控后继续点击；离开目标 package/activity、出现系统设置/支付/授权页时必须停下或人工接管。
真实运行闭环验收
- 启动闭环：脚本版本、配置版本、资源版本、目标 package/activity、权限状态、首屏控件、登录态和停止入口全部记录。
- 动作闭环：每次点击/输入/滑动/OCR 都记录 step、selector/region、bounds、前后截图或业务信号、耗时、重试和结果。
- 异常闭环：权限缺失、控件不存在、页面漂移、网络慢、App 崩溃、脚本异常、系统回收、OCR 低置信分别给恢复路径。
- 发布闭环：单设备灰度、矩阵复测、上一版脚本/配置/资源回滚、运行日志样例、无法验证设备清单和人工接管说明。
- 安全闭环：截图、OCR、日志和导出文件默认脱敏；账号/支付/隐私/风控场景只给合规测试与拒绝口径。
- 版本先钉住：记录 Auto.js、AutoX.js、AutoJS Pro、V8/Rhino/V9 Node 引擎、Android 版本、targetSdk、目标 App 版本、WebView 版本、ADB/打包工具链；版本坑不清只列补证，不直接改 API。
- 设备先钉住：真机品牌、型号、SoC、Android 12/13/14/15、厂商 ROM、分辨率、DPI、横竖屏、刘海/挖孔/导航栏、字体缩放、深色模式、多窗口/折叠屏状态必须进入排障记录。
- 权限先钉住：无障碍服务、悬浮窗权限、截图权限、前台服务通知、自启动、后台弹出界面、电池优化、存储/媒体权限、通知权限逐项确认；权限缺失不是 selector 失效。
- 入口先钉住：脚本入口、目标 package/activity、页面起点、账号/登录态、网络、弹窗、任务来源、停止按钮、回滚方式和人工接管路径必须明确。
- 证据先钉住：每个结论绑定复现步骤、脚本日志、截图/录屏、控件树、selector、bounds、当前包名、当前 Activity、时间戳、设备与系统版本。
- 控件优先：优先使用 id/text/desc/className/packageName/boundsInside/depth/indexInParent 等组合选择器；坐标只做兜底，必须加区域校验和多分辨率适配。
- 等待优先业务信号：点击、滑动、输入、返回后等待控件、文本、Activity、截图特征、列表状态或网络结果；禁止固定 sleep 冒充 waitFor。
- 截图/OCR 只做感知证据：captureScreen、找图、OCR 必须处理授权弹窗、旧帧、旋转、缩放、颜色模式、隐私遮挡；OCR 结果不能替代稳定控件 selector。
- 稳定性日志必须可追溯：记录 step、selector、bounds、action、result、duration、retry、screenshotId、package/activity、battery、foreground/background、异常堆栈。
- 平台差异必须写清：Auto.js 4.x、AutoX.js、AutoJS Pro、不同 JS 引擎、不同 Android API、不同厂商 ROM 的权限、后台、截图、WebView 控件树表现不能混为一谈。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 221 lines · 112 tokens per session scan A 08664349307e
autojs-automation is a skill published in the GitHub repository cass-2003/local-workflow-skill (12 stars, last pushed 2mo ago), licensed MIT. It adds 112 tokens to every session and 7,396 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
orca-emulator-android
Android device and emulator control from inside Orca over adb, with the live device view in Orca's emulator pane. Use when driving an adb-connected emulator or phone on Windows, Linux, or macOS: booting AVDs, taps, swipes, typing, hardware buttons, rotation, app install and launch, runtime permissions, the…
android-tombstone-symbolication
Symbolicate the .NET runtime frames in an Android tombstone file. Extracts BuildIds and PC offsets from the native backtrace, downloads debug symbols from the Microsoft symbol server, and runs llvm-symbolizer to produce function names with source file and line numbers. USE FOR triaging a .NET MAUI or Mono Android app…
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
winapp-maui
Package and sign .NET MAUI Windows apps with winapp, resolving the resizetizer manifest dependency. Use when packaging or signing a .NET MAUI Windows app, building a MAUI MSIX or signed unpackaged build in CI, or fixing 'manifest contains unresolved placeholders ($placeholder$)' errors from winapp package.
apple-search-ads
When the user wants to set up, optimize, or scale Apple Search Ads (ASA) campaigns — including keyword bidding, match types, campaign structure, Creative Product Sets, CPP routing, and ROAS optimization. Use when the user mentions "Apple Search Ads", "ASA", "Search Ads", "Search tab ads", "Today tab ads", "CPT"…
android-pentest
A guide for authorized security testing of Android apps, covering APK inspection, runtime testing, traffic capture, code review, and function hooking. An APK is the installable package used by an Android app.