能力标签
⚙️
Agent工作流

可以穷手机操作器

基于 JavaScript · 无代码搭建完整 AI 自动化流程
英文名:accessibility-agents
⭐ 283 Stars 🍴 27 Forks 💻 JavaScript 📄 MIT 🏷 AI 7.5分
7.5AI 综合评分
workflowa11yaccessibility
✦ AI Skill Hub 推荐

经 AI Skill Hub 精选评估,可以穷手机操作器 获评「推荐使用」。这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 7.5 分,适合有一定技术背景的用户使用。

📚 深度解析
可以穷手机操作器 是一套完整的 AI Agent 自动化工作流方案。随着 AI 能力的不断提升,基于 Agent 的自动化工作流正在成为提升个人和团队效率的核心方式。区别于传统的 RPA 自动化(模拟鼠标键盘操作),AI Agent 工作流通过理解任务意图、动态规划执行路径,能够处理更复杂的非结构化任务。

可以穷手机操作器 工作流的设计遵循"最小配置,最大复用"原则:核心逻辑已经封装好,用户只需配置自己的 API Key 和业务参数即可快速上手。工作流内置错误处理和重试机制,在网络波动或 API 限速等情况下仍能稳定运行,适合作为生产环境的自动化基础设施。

在实际部署时,建议先在测试环境中运行 3-5 次,验证各个环节的输出结果符合预期,再部署到生产环境。AI Skill Hub 评分 7.5 分,是同类 Agent 工作流中的精选推荐。
📋 工具概览

可以穷手机操作器为常用的手机操作器,可以用为Claude为常用的手机操作器。

可以穷手机操作器 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。

GitHub Stars
⭐ 283
开发语言
JavaScript
支持平台
Windows / macOS / Linux
维护状态
轻量级项目,按需更新
开源协议
MIT
AI 综合评分
7.5 分
工具类型
Agent工作流
Forks
27
📖 中文文档
以下内容由 AI Skill Hub 根据项目信息自动整理,如需查看完整原始文档请访问底部「原始来源」。

可以穷手机操作器为常用的手机操作器,可以用为Claude为常用的手机操作器。

可以穷手机操作器 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。

📌 核心特色
  • 可视化 Agent 工作流编排,无需编写复杂代码
  • 支持多步骤自动化任务链,实现全流程无人值守
  • 与外部 API、数据库和第三方服务无缝集成
  • 内置错误处理与自动重试机制,保障稳定运行
  • 提供可复用的自动化模板,快速在同类场景部署
🎯 主要使用场景
  • 自动化日常重复性工作,将精力集中于创造性任务
  • 构建数据采集 → 处理 → 输出的完整自动化管线
  • 实现跨平台、跨系统的数据流转和业务协同
以下安装命令基于项目开发语言和类型自动生成,实际以官方 README 为准。
安装命令
# 方式一:npm 全局安装
npm install -g accessibility-agents

# 方式二:npx 直接运行(无需安装)
npx accessibility-agents --help

# 方式三:项目依赖安装
npm install accessibility-agents

# 方式四:从源码运行
git clone https://github.com/Community-Access/accessibility-agents
cd accessibility-agents
npm install
npm start
📋 安装步骤说明
  1. 访问 GitHub 仓库获取工作流文件
  2. 在对应平台(Dify / Flowise / Make 等)中找到「导入工作流」功能
  3. 上传工作流文件
  4. 按照提示配置必要的环境变量和 API Key
  5. 运行测试确认流程正常后投入使用
以下用法示例由 AI Skill Hub 整理,涵盖最常见的使用场景。
常用命令 / 代码示例
# 命令行使用
accessibility-agents --help

# 基本用法
accessibility-agents [options] <input>

# Node.js 代码中使用
const accessibility_agents = require('accessibility-agents');

const result = await accessibility_agents.run(options);
console.log(result);
以下配置示例基于典型使用场景生成,具体参数请参照官方文档调整。
配置示例
# accessibility-agents 配置说明
# 查看配置选项
accessibility-agents --config-example > config.yml

# 常见配置项
# output_dir: ./output
# log_level: info
# workers: 4

# 环境变量(覆盖配置文件)
export ACCESSIBILITY_AGENTS_CONFIG="/path/to/config.yml"
📑 README 深度解析 真实文档 完整度 63/100 查看 GitHub 原文 →
以下内容由系统直接从 GitHub README 解析整理,保留代码块、表格与列表结构。

Accessibility Agents

License: MIT GitHub release GitHub stars GitHub contributors WCAG 2.2 AA

AI and automated tools are not perfect. They miss things, make mistakes, and cannot replace testing with real screen readers and assistive technology. Always verify with VoiceOver, NVDA, JAWS, and keyboard-only navigation. This tooling is a helpful starting point, not a substitute for real accessibility testing.

A community-driven open-source project automating accessibility, efficiency, and productivity through AI-based agents, skills, custom instructions, and prompts.

A sincere thanks goes out to Taylor Arndt and Jeff Bishop for leading the charge in building this community project. It started because LLMs consistently forget accessibility - skills get ignored, instructions drift out of context, ARIA gets misused, focus management gets skipped, color contrast fails silently. They got tired of fighting it and built an agent team that will not let it slide. Now we want to make more magic together.

We want more contributors! If you care about making software accessible to blind and low vision users, please consider submitting a PR. Every improvement to these agents helps developers ship more inclusive software for the people who need it most.

About This Project

Accessibility Agents was founded by Taylor Arndt (COO at Techopolis) and Jeff Bishop because accessibility is how they work, not something bolted on at the end. When AI coding tools consistently failed at accessibility, they built the team they wished existed - and opened it to the world.

This is a community project. The more perspectives, lived experiences, and domain knowledge that go into it, the better it serves the blind and low vision community. If you have ideas, open a discussion. If you have fixes, open a PR. Every contribution matters.

System Requirements

⚠️ CRITICAL: To remain current with Accessibility Agents and ensure proper functionality, you must keep all tools updated to their latest versions. New platform capabilities, API changes, accessibility features, and bug fixes directly impact agent behavior.

Required Tools (Latest Versions)

For GitHub Copilot (VS Code):

  • VS Code: Latest stable release and/or VS Code Insiders (Download)
  • GitHub Copilot Extension: Latest version from VS Code Marketplace
  • GitHub Copilot Chat Extension: Latest version from VS Code Marketplace
  • Node.js: v18.0.0 or higher (for CLI tools like axe-core)

For Claude Code:

  • Claude Code CLI: Latest version (Installation)
  • Claude Subscription: Pro, Max, or Team plan

For Gemini CLI:

Operating Systems:

  • macOS: 10.15 (Catalina) or later
  • Windows: Windows 10/11 with PowerShell 5.1+ (pre-installed)

Extensions → @installed → Click update icon next to GitHub Copilot extensions

5.0 Installation Direction

Version 5.0.0 moves Accessibility Agents to a GitHub Skills installation flow backed by native Go binaries for setup, health checks, repair, and hook management.

Planned 5.0 flow:

gh skill install Community-Access/accessibility-agents
gh skill setup Community-Access/accessibility-agents
gh skill health Community-Access/accessibility-agents

Important:

  • gh skill install is the future primary install path.
  • The setup tooling is being implemented as native Go binaries, not Node.js scripts.
  • Node.js is still required for the MCP server itself, but not for the installer experience.

Build the 5.0 CLI

If you are contributing to the 5.0 installer transition or preparing release artifacts, install Go and build the native CLI locally.

Windows:

winget install --id GoLang.Go --exact --accept-package-agreements --accept-source-agreements
go version
pwsh -NoProfile -File scripts/build-go-cli.ps1

macOS:

brew install go
go version
bash scripts/build-go-cli.sh

The compiled binaries are written to go-cli/bin/. Windows outputs are .exe files. macOS outputs are native CLI executables.

GitHub Actions also builds the Go CLI automatically on Windows, macOS, and Linux via .github/workflows/build-go-cli.yml.

5.0+ uses the GitHub Skills / Go CLI path as the supported installer path.

Install

Use the GitHub Skills installer:

gh skill install Community-Access/accessibility-agents

Then run setup/health/repair utilities as needed:

gh skill setup Community-Access/accessibility-agents
gh skill health Community-Access/accessibility-agents
gh skill repair Community-Access/accessibility-agents

Uninstall

gh skill uninstall Community-Access/accessibility-agents

If you need manual cleanup details, see UNINSTALL.md.

Safe installation — your files are never overwritten

The setup flow is designed to be additive and non-destructive:

  • Agent files (~/.claude/agents/, .github/agents/) - existing files are skipped, not replaced. A message tells you which agents were skipped so you know what you already have.
  • Config files (copilot-instructions.md, copilot-review-instructions.md, copilot-commit-message-instructions.md) - our content is wrapped in `` markers and merged into your existing file. Your content above and below the markers is always preserved. If the file does not exist, it is created.
  • Asset directories (skills/, instructions/, prompts/) - copied file-by-file; files that already exist are skipped.
  • Manifest file (.a11y-agent-manifest) - tracks every file we installed. Repair and update flows use this list to ensure they only touch files we own, never user-created agents. When contributors add new agents to the repo, those files are installed on the next setup/repair pass and added to the manifest.

Updates are equally safe - maintenance flows never delete user agent files. If a file is not in the manifest (meaning you created it yourself), it will not be modified or removed.

To reinstall a specific agent from scratch, delete it first and rerun setup or repair.

Post-Install Validation

After installation, run a validation and self-repair pass. This verifies that all installed surfaces are intact, MCP base dependencies are present, and Playwright is functional if installed.

You can also run it manually at any time:

PowerShell (Windows) — validate only:

pwsh -NoProfile -ExecutionPolicy Bypass -File scripts/repair-install.ps1 -SummaryPath .a11y-agent-team-install-summary.json

PowerShell (Windows) — validate and auto-repair:

pwsh -NoProfile -ExecutionPolicy Bypass -File scripts/repair-install.ps1 -SummaryPath .a11y-agent-team-install-summary.json -Repair

Shell (macOS / Linux / Git Bash) — validate only:

bash scripts/repair-install.sh --summary=.a11y-agent-team-install-summary.json --validate-only

Shell (macOS / Linux / Git Bash) — validate and auto-repair:

bash scripts/repair-install.sh --summary=.a11y-agent-team-install-summary.json

VS Code users can run these via the Terminal > Run Task menu: look for the Repair: task group.

The repair pass checks and fixes:

  • Destination paths — every installed surface (Claude, Copilot, Codex, Gemini, MCP) exists on disk
  • MCP base dependencies@modelcontextprotocol/sdk and zod are present; runs npm install --omit=dev if missing
  • Playwrightplaywright-core is installed; Chromium is functional; re-runs setup if broken
  • Copilot profile root cleanup — removes stray agent/prompt/instruction files placed directly in VS Code profile roots instead of the expected subdirectories

Findings are appended to the install summary JSON and a separate .a11y-agent-team-repair-summary.json is written next to it. See Troubleshooting for symptom-specific guidance.

GitHub Skills Rollout Guides

For the GitHub Skills specification rollout and release-readiness workflows, see:

  • docs/guides/GITHUB-SKILLS-CLI-READINESS.md
  • docs/guides/SKILLS-RELEASE-READINESS-TEST-PLAN.md
  • .github/workflows/skills-cli-readiness.yml
  • .github/workflows/skills-release-readiness.yml

Quick Start

Advanced Guides

The following guides cover advanced configuration, cross-platform handoff, and distribution.

GuideWhat It Covers
[Cross-Platform Handoff](docs/advanced/cross-platform-handoff.md)Seamless handoff between Claude Code and Copilot
[Advanced Scanning Patterns](docs/advanced/advanced-scanning-patterns.md)Background scanning, worktree isolation, large libraries
[Metadata and Markup Conventions](docs/guides/metadata-markup-conventions.md)Recommended metadata fields and markdown structure conventions for agents, skills, and instructions
[Plugin Packaging](docs/advanced/plugin-packaging.md)Packaging and distributing agents for different environments
[Platform References](docs/advanced/platform-references.md)External documentation sources with feature-to-source mapping
[Playwright High-Impact Checks](docs/guides/playwright-high-impact-checks.md)High-impact runtime accessibility checks for keyboard traps, serious violations, and viewport risk
[CI Integrity Guards](docs/guides/ci-integrity-guards.md)Guard rails for workflow invariants, config/schema drift, and documentation version pin checks
[Release Communications Checklist](docs/guides/release-communications-checklist.md)Repeatable release messaging checklist for changelog, notes, publish, and announcements
[Experimental Codex Multi-Agent Roles](docs/guides/codex-experimental-multi-agent.md)Optional TOML-based Codex roles for focused accessibility passes
[Research Sources](docs/RESEARCH-SOURCES.md)Authoritative sources (W3C APG, WebAIM, WCAG 2.2, Deque) that informed every agent rule

Example Project

The example/ directory contains a deliberately broken web page with 20+ intentional accessibility violations. Use it to practice with the agents and see how they catch real issues. See the example README for details.

Help → Check for Updates (or auto-updates if enabled in settings)

Optional Customization

Update GitHub Copilot Extensions

GitHub Workflow Agents

The following agents handle GitHub repository management, triage, and workflow automation.

AgentRole
**github-hub**Orchestrator. Routes GitHub management tasks to the right specialist from plain English.
**daily-briefing**Morning overview - open issues, PR queue, CI status, security alerts in one report.
**pr-review**PR diff analysis with confidence per finding, delta tracking, and inline comments.
**issue-tracker**Issue triage - priority scoring, duplicate detection, action inference, project board sync.
**analytics**Repository health scoring (0-100/A-F), velocity metrics, bottleneck detection.
**insiders-a11y-tracker**Track accessibility changes in VS Code Insiders and custom repos with WCAG mapping.
**repo-admin**Collaborator management, branch protection rules, access audits.
**team-manager**Onboarding, offboarding, org team membership, permission management.
**contributions-hub**Discussions, community health metrics, first-time contributor insights.
**template-builder**Guided wizard for issue/PR/discussion templates - no YAML knowledge required.
**repo-manager**Repository scaffolding - labels, CI, CONTRIBUTING, SECURITY, issue templates.

See the Agent Reference Guide for deep dives on every agent, example prompts, behavioral constraints, and instructor-led walkthroughs.

GitHub Workflow Docs

The following guide covers all GitHub workflow agents and their invocation syntax.

GuideWhat It Covers
[GitHub Workflow Agents](docs/agents/README.md#github-workflow-agents)All 10 workflow agents with invocation syntax, examples, and instructor-led walkthroughs

Update VS Code

VS Code 1.113 Highlights

VS Code 1.113 is currently the most relevant baseline for GitHub Copilot users of this repo. The release added several changes that improve Accessibility Agents workflows directly:

  • MCP across agent types - MCP servers configured in VS Code now bridge into Copilot CLI and Claude agents, which makes this repo's MCP guidance more consistent across local, CLI, and Claude workflows.
  • Chat Customizations editor - Chat: Open Chat Customizations gives you one place to inspect and manage instructions, prompt files, agents, skills, MCP servers, and plugins.
  • Broader Agent Debug coverage - Agent Debug Logs now cover Copilot CLI and Claude agent sessions in addition to local sessions.
  • Integrated browser improvements - local HTTPS testing with self-signed certificates is easier, and browser tab management is better for accessibility testing workflows.
  • Nested subagents - VS Code now supports nested subagent delegation. For this repo, that is treated as an optional platform capability, not a default architecture choice. We favor explicit coordinator-worker flows and keep nested subagents disabled by default.

Subagent stance for this repo:

  • Reward: bounded subagents improve specialization, parallel analysis, and audit structure.
  • Risk: unrestricted or nested subagents can increase duplicate findings, wrong-agent selection, token cost, and debugging complexity.
  • Recommendation: use explicit, allowlisted specialist delegation; avoid recursive subagent chains unless a workflow is intentionally designed for them.

For official details, see the VS Code 1.113 release notes: https://code.visualstudio.com/updates/v1_113.

VS Code 1.113 Workflow Tips

If you are using Accessibility Agents in VS Code 1.113, these settings and commands are the most useful starting point:

{
  "chat.useCustomizationsInParentRepositories": true,
  "github.copilot.chat.agentDebugLog.enabled": true,
  "github.copilot.chat.agentDebugLog.fileLogging.enabled": true,
  "chat.imageSupport.enabled": true
}
  • Run Chat: Open Chat Customizations to inspect loaded instructions, agents, skills, MCP servers, and plugins.
  • Use /troubleshoot when a customization is not loading or an expected tool is missing.
  • Keep chat.subagents.allowInvocationsFromSubagents disabled unless you are intentionally experimenting with recursive orchestration patterns.

That last point is a repo recommendation, not a VS Code limitation.

Custom Thinking Phrases (VS Code 1.110+)

VS Code users: Personalize the loading text that appears while agents think with accessibility-themed phrases.

Add to VS Code Settings (settings.json):

{
  "chat.agent.thinking.phrases": {
    "mode": "append",  // Adds to default phrases
    "phrases": [
      "Checking contrast ratios...",
      "Testing with screen readers...",
      "Verifying keyboard navigation...",
      "Reviewing ARIA patterns...",
      "Scanning for accessibility barriers...",
      "Consulting WCAG 2.2..."
    ]
  }
}

Options:

  • "mode": "append" - Adds your phrases to VS Code's default list (recommended)
  • "mode": "replace" - Only shows your custom phrases

Why This Matters:

  • Reinforces accessibility focus during agent work
  • Reminds team members that accessibility is actively considered
  • Optional fun enhancement to make wait time more engaging

How to Add:

  1. Open VS Code Settings (Ctrl/Cmd + ,)
  2. Click "Open Settings (JSON)" icon in top-right
  3. Add the chat.agent.thinking.phrases setting
  4. Reload window (Command Palette → "Developer: Reload Window")

Community Contributions: Have a great accessibility-themed thinking phrase? Submit a PR to add it to our recommended list in CONTRIBUTING.md!

Install from VS Code Marketplace (Coming Soon)

GitHub Copilot users (VS Code): A VS Code extension is in development that will allow one-click install from the Extensions marketplace. Until then, use the one-liner install above or the manual steps in Getting Started.

What the extension will include:

  • ✅ 59 fully-integrated agents in GitHub Copilot (VS Code and CLI)
  • ✅ 18 reusable accessibility skills (WCAG rules, severity scoring, scanning patterns)
  • ✅ 106 custom prompts for web audits, document audits, GitHub workflows, and developer tooling
  • ✅ 6 workspace instructions (automatic WCAG AA enforcement on every chat)
  • ✅ 100% source citation coverage (all agents cite authoritative standards)
  • ✅ Auto-update mechanism (new agents and features arrive automatically)

For other platforms (Claude Code, Gemini, Claude Desktop, Codex), see Getting Started.

🎯 aiskill88 AI 点评 A 级 2026-05-26

可以穷手机操作器为常用的手机操作器,可以用为Claude为常用的手机操作器。不得载的题。

⚡ 核心功能
👥 适合人群
自动化工程师和运维人员项目经理和业务分析师希望减少重复性工作的专业人士数字化转型团队
🎯 使用场景
  • 自动化日常重复性工作,将精力集中于创造性任务
  • 构建数据采集 → 处理 → 输出的完整自动化管线
  • 实现跨平台、跨系统的数据流转和业务协同
⚖️ 优点与不足
✅ 优点
  • +MIT 协议,可免费商用
  • +大幅减少重复性人工操作
  • +可视化流程,清晰直观
  • +可扩展性强,支持复杂场景
⚠️ 不足
  • 初始配置和调试需投入一定时间
  • 强依赖外部服务的稳定性
  • 复杂场景需具备一定技术基础
⚠️ 使用须知

AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。

建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。

📄 License 说明

✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。

🔗 相关工具推荐
🧩 你可能还需要
基于当前 Skill 的能力图谱,自动补全的工具组合
❓ 常见问题 FAQ
accessibility-agents 是一款JavaScript开发的AI辅助工具。开源AI工作流:Accessibility review agents for Claude Code, GitHub Copilot, and Claude Desktop.。⭐283 · JavaScript 主要应用场景包括:可以用为手机操作器的穷手机操作器、手机操作器的穷手机操作器、Claude为常用的手机操作器。。
💡 AI Skill Hub 点评

AI Skill Hub 点评:可以穷手机操作器 的核心功能完整,质量良好。对于自动化工程师和运维人员来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。

⬇️ 获取与下载
⬇ 下载源码 ZIP

✅ MIT 协议 · 可免费商用 · 直接从 aiskill88 服务器下载,无需跳转 GitHub

📚 深入学习 可以穷手机操作器
查看分步骤安装教程和完整使用指南,快速上手这款工具
🌐 原始信息
原始名称 accessibility-agents
原始描述 开源AI工作流:Accessibility review agents for Claude Code, GitHub Copilot, and Claude Desktop.。⭐283 · JavaScript
Topics workflowa11yaccessibility
GitHub https://github.com/Community-Access/accessibility-agents
License MIT
语言 JavaScript
🔗 原始来源
🐙 GitHub 仓库  https://github.com/Community-Access/accessibility-agents 🌐 官方网站  https://github.com/taylorarndt/a11y-agent-team#readme

收录时间:2026-05-26 · 更新时间:2026-05-26 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。