Token导航 LogoToken导航TokenDH.com
运维和基础设施external-servicegithub未标认证来源可访问clear审计提醒

remote-work远程工作

Agent Skill

remote-work 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

1,958

周安装

99

GitHub Stars

55

下载量

784
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:remote-work(远程工作)
来源仓库:https://github.com/rysweet/amplihack
仓库路径:skills/remote-work
安装命令:
npx skills add https://github.com/rysweet/amplihack --skill remote-work
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/rysweet/amplihack --skill remote-work

简介

用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态或协作事项进行整理。
  • 通过 npx skills add 命令从指定仓库安装并使用。
  • 建议确认权限范围和维护状态,避免触发不必要的联网或文件操作。
  • remote-work 属于运维和基础设施类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Remote Work Skill

Execute amplihack tasks on remote Azure VMs using the /amplihack:remote command.

When to Use

Use this skill when you want to:

  • Run long-running tasks on remote VMs
  • Leverage more powerful Azure compute
  • Distribute work across multiple machines
  • Isolate experimental or risky work
  • Work in specific Azure regions

How It Works

When you say things like:

  • "Run this remotely"
  • "Execute on an eastus VM"
  • "Offload this to Azure"
  • "Use a remote machine for this"

I will:

  1. Extract your task description
  2. Detect region preferences (if mentioned)
  3. Recommend appropriate VM size based on task complexity
  4. Execute /amplihack:remote with proper options
  5. Guide you on accessing results

Region Detection

I automatically detect Azure regions from your request:

  • "eastus", "east us" → --region eastus
  • "westus", "west us" → --region westus
  • "centralus" → --region centralus
  • (and other Azure regions)

VM Size Recommendations

Based on your task complexity keywords:

Small (size s): Quick analysis, simple fixes

  • Keywords: "quick", "simple", "small", "analyze"
  • Size: s (8GB RAM)

Medium (size m): Standard development work

  • Keywords: "implement", "feature", "refactor"
  • Size: m (64GB RAM) - default

Large (size l): Complex refactoring, large codebases

  • Keywords: "large", "comprehensive", "complex", "entire codebase"
  • Size: l (128GB RAM)

Extra Large (size xl): Intensive compute tasks

  • Keywords: "intensive", "heavy", "massive"
  • Size: xl (256GB RAM)

Instructions

When activated:

  1. Parse the user's request to extract:

- The task description - Region preference (if mentioned) - Complexity hints for VM sizing

  1. Determine VM size based on keywords:

- Default to "m" unless keywords suggest otherwise - Use "s" for simple/quick tasks - Use "l" for large/complex tasks - Use "xl" for intensive/massive tasks

  1. Construct the command: /amplihack:remote --region {region} --vm-size {size} auto "{task}" If no region mentioned, omit --region (uses azlin default)
  2. Execute the command using the SlashCommand tool
  3. Guide the user on next steps:

- Results will be in ~/.amplihack/.claude/runtime/logs/remote/ - Git branches in remote-exec/ namespace - How to view logs and integrate changes

Examples

Example 1: Simple with Region

User: "Run this analysis remotely on an eastus VM"

Your Response:

I'll execute this analysis on a remote Azure VM in eastus.

Detected:
- Task: analysis work
- Region: eastus
- Size: s (small - suitable for analysis)

Executing: /amplihack:remote --region eastus --vm-size s auto "analysis work"

Then use SlashCommand tool to execute it.

Example 2: Large Refactoring

User: "I need to offload this comprehensive refactoring to Azure, use westus2"

Your Response:

I'll execute this comprehensive refactoring on a remote Azure VM in westus2.

Detected:
- Task: comprehensive refactoring
- Region: westus2
- Size: l (large - for comprehensive work)

Executing: /amplihack:remote --region westus2 --vm-size l auto "comprehensive refactoring"

Example 3: No Region Specified

User: "Run this remotely"

Your Response:

I'll execute this task on a remote Azure VM using your default region.

Detected:
- Task: [extract from context]
- Region: default (from azlin config)
- Size: m (medium - standard)

Executing: /amplihack:remote --vm-size m auto "[task]"

Result Integration

After execution completes, guide the user:

Remote execution complete!

Results:
- Logs: .claude/runtime/logs/remote/
- Branches: Check `git branch -r | grep remote-exec`
- Integration: Review changes and merge if ready

To view logs:
  cat .claude/runtime/logs/remote/remote-exec-*.log

To merge remote changes:
  git fetch
  git merge remote-exec/main

Error Handling

If the command fails, explain common issues:

Secrets Detected: "Remote execution blocked - secrets found in repository. Please remove hardcoded secrets and use.env files instead."

Azlin Not Configured: "Azlin is not configured. Please run: azlin configure"

Timeout: "Task timed out. The VM has been preserved for inspection. Use azlin connect <vm-name> to investigate."

Best Practices

Remind users:

  • Keep git state clean before remote execution
  • Use .env for secrets (never hardcode)
  • Monitor Azure costs with azlin list
  • Clean up VMs after debugging with azlin kill <vm-name>

Token Budget

  • Core instructions: ~1,200 tokens
  • Examples and guidance: ~400 tokens
  • Total: ~1,600 tokens (well under 2,000 limit)

See Also

  • /amplihack:remote - Direct command interface
  • Azlin documentation: https://github.com/rysweet/azlin
  • Remote execution module: ~/.amplihack/.claude/tools/amplihack/remote/README.md

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

25.58%
按下载量换算201

Antigravity

24.87%
按下载量换算195

OpenCode

17.88%
按下载量换算140

Gemini CLI

10.91%
按下载量换算86

windsurf

7.17%
按下载量换算56

Cursor

3.43%
按下载量换算27

安全审计

Gen Agent Trust Hub

通过

Socket

可疑

Snyk

通过

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills