remote-work

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Remote Work Skill

远程工作技能

Execute amplihack tasks on remote Azure VMs using the

/amplihack:remote

command.

使用

/amplihack:remote

命令在远程Azure VM上执行amplihack任务。

When to Use

使用场景

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

当你需要以下操作时，可使用此技能：

在远程VM上运行耗时较长的任务
利用更强大的Azure计算资源
在多台机器上分配工作负载
隔离实验性或有风险的工作
在特定Azure区域中工作

How It Works

工作原理

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:

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

当你说出类似以下内容时：

"远程运行此任务"
"在eastus的VM上执行"
"将此任务卸载到Azure"
"使用远程机器处理此任务"

我会：

提取你的任务描述
检测区域偏好（如果有提及）
根据任务复杂度推荐合适的VM规格
使用正确的选项执行
```
/amplihack:remote
```
指导你如何访问结果

Region Detection

区域检测

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)

我会自动从你的请求中检测Azure区域：

"eastus", "east us" → --region eastus
"westus", "west us" → --region westus
"centralus" → --region centralus
(以及其他Azure区域)

VM Size Recommendations

VM规格推荐

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)

根据任务复杂度关键词：

小型（规格s）：快速分析、简单修复

关键词："quick", "simple", "small", "analyze"
规格：s (8GB RAM)

中型（规格m）：标准开发工作

关键词："implement", "feature", "refactor"
规格：m (64GB RAM) - 默认

大型（规格l）：复杂重构、大型代码库

关键词："large", "comprehensive", "complex", "entire codebase"
规格：l (128GB RAM)

超大型（规格xl）：密集型计算任务

关键词："intensive", "heavy", "massive"
规格：xl (256GB RAM)

Instructions

操作步骤

When activated:

Parse the user's request to extract:
- The task description
- Region preference (if mentioned)
- Complexity hints for VM sizing
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
Construct the command:
```
/amplihack:remote --region {region} --vm-size {size} auto "{task}"
```
If no region mentioned, omit --region (uses azlin default)
Execute the command using the SlashCommand tool
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

激活技能后：

解析用户请求以提取：
- 任务描述
- 区域偏好（如果提及）
- 用于VM规格选择的复杂度提示
根据关键词确定VM规格：
- 默认使用"m"，除非关键词提示其他规格
- 简单/快速任务使用"s"
- 大型/复杂任务使用"l"
- 密集型/大规模任务使用"xl"
构造命令：
```
/amplihack:remote --region {region} --vm-size {size} auto "{task}"
```
如果未提及区域，则省略--region（使用azlin默认值）
使用SlashCommand工具执行命令
指导用户后续步骤：
- 结果将存储在
```
~/.amplihack/.claude/runtime/logs/remote/
```
- Git分支位于
```
remote-exec/
```
  命名空间下
- 如何查看日志和整合更改

Examples

示例

Example 1: Simple with Region

示例1：指定区域的简单任务

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.

用户："在eastus的VM上远程运行此分析"

你的回复：

我将在eastus区域的远程Azure VM上执行此分析任务。

检测到：
- 任务：分析工作
- 区域：eastus
- 规格：s（小型 - 适合分析任务）

正在执行：/amplihack:remote --region eastus --vm-size s auto "分析工作"

然后使用SlashCommand工具执行该命令。

Example 2: Large Refactoring

示例2：大型重构任务

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"

用户："我需要将这个全面的重构任务卸载到Azure，使用westus2"

你的回复：

我将在westus2区域的远程Azure VM上执行这个全面的重构任务。

检测到：
- 任务：全面重构
- 区域：westus2
- 规格：l（大型 - 适用于全面工作）

正在执行：/amplihack:remote --region westus2 --vm-size l auto "全面重构"

Example 3: No Region Specified

示例3：未指定区域

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]"

用户："远程运行此任务"

你的回复：

我将在远程Azure VM上执行此任务，使用你的默认区域。

检测到：
- 任务：[从上下文中提取]
- 区域：默认（来自azlin配置）
- 规格：m（中型 - 标准规格）

正在执行：/amplihack:remote --vm-size m auto "[task]"

Result Integration

结果整合

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

执行完成后，指导用户：

远程执行完成！

结果信息：
- 日志：.claude/runtime/logs/remote/
- 分支：查看`git branch -r | grep remote-exec`
- 整合：检查更改并在准备就绪后合并

查看日志的命令：
  cat .claude/runtime/logs/remote/remote-exec-*.log

合并远程更改的命令：
  git fetch
  git merge remote-exec/main

Error Handling

错误处理

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."

如果命令执行失败，解释常见问题：

检测到密钥： "远程执行已阻止 - 在仓库中检测到密钥。请移除硬编码的密钥，改用.env文件。"

Azlin未配置： "Azlin未配置。请运行：

azlin configure

超时： "任务超时。VM已保留以供检查。使用

azlin connect <vm-name>

进行调查。"

Best Practices

最佳实践

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>
```

提醒用户：

远程执行前保持Git状态干净
使用
```
.env
```
存储密钥（切勿硬编码）
使用
```
azlin list
```
监控Azure成本
调试完成后使用
```
azlin kill <vm-name>
```
清理VM

Token Budget

Token预算

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

核心指令：约1,200 tokens
示例和指导：约400 tokens
总计：约1,600 tokens（远低于2,000限制）

另请参阅

```
/amplihack:remote
```
- Direct command interface
Azlin documentation: https://github.com/rysweet/azlin

Remote execution module:

~/.amplihack/.claude/tools/amplihack/remote/README.md

```
/amplihack:remote
```
- 直接命令接口
Azlin文档：https://github.com/rysweet/azlin

远程执行模块：

~/.amplihack/.claude/tools/amplihack/remote/README.md

remote-work

Original

Translation

Remote Work Skill

远程工作技能

When to Use

使用场景

How It Works

工作原理

Region Detection

区域检测

VM Size Recommendations

VM规格推荐

Instructions

操作步骤

Examples

示例

Example 1: Simple with Region

示例1：指定区域的简单任务

Example 2: Large Refactoring

示例2：大型重构任务

Example 3: No Region Specified

示例3：未指定区域

Result Integration

结果整合

Error Handling

错误处理

Best Practices

最佳实践

Token Budget

Token预算

See Also

另请参阅