Token导航 LogoToken导航TokenDH.com
开发需要联网clawhub未标认证来源可访问clear审计提醒

yuyonghao-code-sandbox余永浩代码沙箱

Agent Skill

yuyonghao-code-sandbox 用于辅助 Python 项目开发、测试和数据处理,适合在 OpenClaw 中需要阅读 Python 代码、运行测试或整理脚本流程时使用。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

3,998

周安装

170

GitHub Stars

公开资料未说明

下载量

1,401
OpenClaw

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

2

许可证

MIT-0

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:yuyonghao-code-sandbox(余永浩代码沙箱)
来源仓库:https://github.com/yuyonghao-123/yuyonghao-code-sandbox
安装命令:
openclaw skills install yuyonghao-code-sandbox
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。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install yuyonghao-code-sandbox

简介

用于执行 Node.js、Python、Go 和 Rust 代码的安全沙箱,具有超时、CPU 和隔离的临时目录约束。

SKILL.md

Code Sandbox Skill

🔒 Secure code execution sandbox for OpenClaw agents.

Features

  • Multi-language Support: Node.js, Python, Go, Rust
  • Resource Limits: Timeout, memory, CPU constraints
  • Isolation: Temporary directories, process isolation
  • Execution History: Track all executions with metrics
  • Safe Cleanup: Automatic cleanup of temporary files

Installation

# Already in workspace skills directory
cd skills/code-sandbox
npm install

Usage

JavaScript API

const { CodeSandbox } = require('./src/sandbox');

const sandbox = new CodeSandbox({
  defaultTimeout: 30000,      // 30 seconds
  defaultMemoryLimit: 512,    // 512 MB
});

// Execute Node.js code
const result = await sandbox.execute({
  language: 'node',
  code: 'console.log("Hello World!");',
});

console.log(result);
// {
//   success: true,
//   output: "Hello World!\
",
//   error: "",
//   exitCode: 0,
//   executionTime: 45,
//   memoryUsed: 0
// }

// Execute Python code
const pythonResult = await sandbox.execute({
  language: 'python',
  code: 'print("Hello from Python!")',
  config: {
    timeout: 5000,  // 5 seconds
  }
});

CLI Demo

# Run demo
npm run demo

# Run tests
npm test

Supported Languages

LanguageVersionDescription
node20.xNode.js JavaScript runtime
python3.11Python interpreter
go1.21Go compiler and runtime
rust1.75Rust compiler (rustc)

Configuration

const config = {
  defaultTimeout: 30000,      // Default timeout in ms
  defaultMemoryLimit: 512,    // Default memory limit in MB
  defaultCpuLimit: 1,         // Default CPU cores
  tmpDir: '/path/to/tmp',     // Temporary directory for execution
};

const sandbox = new CodeSandbox(config);

Per-execution Config

await sandbox.execute({
  language: 'node',
  code: '...',
  config: {
    timeout: 10000,      // Override timeout
    memoryLimit: 256,    // Override memory limit
  }
});

Security Features

Current Implementation (v0.1.0)

  • ✅ Process isolation (separate processes)
  • ✅ Temporary directory isolation
  • ✅ Timeout enforcement
  • ✅ Output buffering limits (10MB)
  • ✅ Automatic cleanup

Planned (v0.2.0+)

  • ⏳ Docker containerization
  • ⏳ Windows Job Objects for resource limits
  • ⏳ Network isolation
  • ⏳ Filesystem read-only mode
  • ⏳ Seccomp profiles (Linux)
  • ⏳ Dangerous operation filtering

Execution History

// Get last 100 executions
const history = sandbox.getHistory(100);

console.log(history);
// [
//   {
//     executionId: "exec_1710691200000_a1b2c3d4",
//     timestamp: "2026-03-17T09:00:00.000Z",
//     language: "node",
//     codeLength: 42,
//     success: true,
//     executionTime: 45,
//     memoryUsed: 0
//   },
//   ...
// ]

Error Handling

try {
  const result = await sandbox.execute({
    language: 'node',
    code: 'throw new Error("Oops!");',
  });
  
  if (!result.success) {
    console.error('Execution failed:', result.error);
  }
} catch (error) {
  console.error('Sandbox error:', error.message);
}

Testing

# Run all tests
npm test

# Test specific language
node test/node-test.js
node test/python-test.js

Limitations

Current Version (v0.1.0)

  • No Docker isolation (uses process-level isolation only)
  • No memory limit enforcement (relies on OS)
  • No network isolation
  • Memory usage tracking not implemented

Known Issues

  • Windows path handling may need adjustment
  • Rust/Go compilation errors could be clearer
  • Large outputs may be truncated

Roadmap

v0.2.0 (Next Release)

  • [ ] Docker containerization support
  • [ ] Windows Job Objects integration
  • [ ] Memory limit enforcement
  • [ ] Network isolation (disable by default)

v0.3.0

  • [ ] Seccomp profiles (Linux)
  • [ ] Filesystem whitelist/blacklist
  • [ ] Code scanning for dangerous patterns
  • [ ] Concurrent execution limits

v0.4.0

  • [ ] REST API server
  • [ ] WebSocket streaming output
  • [ ] Batch execution support
  • [ ] Prometheus metrics

Safety Warnings

⚠️ This sandbox provides basic isolation but is NOT suitable for:

  • Running untrusted code from strangers
  • Production environments without additional security
  • Executing code that requires high security guarantees

For production use, implement:

  • Docker containers with security profiles
  • Network isolation
  • Resource quotas
  • Code scanning and validation

License

MIT License - See LICENSE file for details

Contributing

Contributions welcome! Please read CONTRIBUTING.md first.


Version: 0.1.0 Last Updated: 2026-03-17 Author: OpenClaw Team

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

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

OpenClaw

75.9%
按下载量换算1,063

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills