Back to Skills

agentbay-aio-skills

agentbay-ai
Updated 2 days ago
6 views
40
2
40
View on GitHub
Metaaitesting

About

This skill executes code in a secure sandbox environment when users request to run, execute, or evaluate code snippets. It supports multiple languages including Python, JavaScript, R, and Java through the AgentBay SDK's run_code function. Developers should use it for any code execution tasks like running scripts, plotting graphs, or checking output results.

Quick Install

Claude Code

Recommended
Primary
npx skills add agentbay-ai/agentbay-skills -a claude-code
Plugin CommandAlternative
/plugin add https://github.com/agentbay-ai/agentbay-skills
Git CloneAlternative
git clone https://github.com/agentbay-ai/agentbay-skills.git ~/.claude/skills/agentbay-aio-skills

Copy and paste this command in Claude Code to install this skill

Documentation

AgentBay AIO 技能

触发条件:何时使用本技能

当用户意图运行或执行代码时即应使用本技能,即使用户没有说「沙箱」。典型表述(遇到即触发):

  • 英文:run this code、execute this script、run the code、execute the script、plot with Python、draw a graph、evaluate this snippet、what's the output of this code、can you run this
  • 中文:帮我跑一下这段代码、运行这段代码、执行这个脚本、用 Python 画个图、跑一下、看下运行结果、执行一下、帮我运行

依赖与环境

本技能依赖 wuying-agentbay-sdk 包。

最佳实践(Agent 必读):

为避免污染用户当前项目的环境,请务必使用技能目录下的独立虚拟环境

  1. 检查/创建虚拟环境

    # 检查是否存在 .venv,不存在则创建
    if [ ! -d ".venv" ]; then python3 -m venv .venv; fi
    
  2. 激活并安装依赖(仅首次或缺失时安装,使用阿里云镜像):

    source .venv/bin/activate
    pip show wuying-agentbay-sdk >/dev/null 2>&1 || pip install wuying-agentbay-sdk -i https://mirrors.aliyun.com/pypi/simple/
    
  3. 运行脚本(在虚拟环境中):

    # 确保在 venv 下运行
    python scripts/run_code.py --code "..." --language python
    

(若用户明确要求在全局环境运行,或当前已在特定环境中,可跳过创建 venv,但仍建议使用镜像源安装。)

API Key(仅报错时引导)

默认先直接执行脚本,不要每次运行前都提醒用户配置 API Key。 仅当脚本报错「Missing API key」或类似信息时,再引导用户配置:

  1. AgentBay 控制台 申请 API Key。
  2. 在本地创建配置文件(脚本会从该路径读取):
    • macOS / Linux~/.config/agentbay/api_key(或 $XDG_CONFIG_HOME/agentbay/api_key
    • Windows%USERPROFILE%\.config\agentbay\api_key
  3. 示例(macOS/Linux):mkdir -p ~/.config/agentbay,再将密钥写入 ~/.config/agentbay/api_key。Windows:在 %USERPROFILE%\.config\agentbay 下创建名为 api_key 的无扩展名文件。
  4. 告知用户完成后重试,不要只复述报错。

使用场景

所有代码执行类请求均默认使用本技能。 只要用户要求运行代码、执行脚本、评估片段或查看代码输出(Python、JavaScript、R 或 Java),即使用本技能——即使用户未提「沙箱」。使用 code_latest 创建 AgentBay 沙箱,通过 run_code 执行代码并返回结果。在本技能可用时,不要在本地终端执行用户代码;应使用沙箱以保证隔离与行为一致。

概述

使用 AgentBay SDK 以 code_latest 镜像创建会话,在沙箱中通过 run_code 执行代码,解析 EnhancedCodeExecutionResult,并销毁会话。

执行方式(Agent 必读)

请通过本技能自带的脚本执行用户代码 在技能目录下执行:

python scripts/run_code.py --code "用户要执行的代码" --language python

从文件执行:

python scripts/run_code.py --code-file /path/to/file.py --language python

需要结构化输出时加 --json。脚本会从配置文件或环境变量读取 API Key,创建沙箱、执行并返回结果。

(若用户在自己的 Python 项目中集成 AgentBay,可参考 wuying-agentbay-sdk 的同步/异步用法;Agent 执行本技能时仅需调用上述脚本。)

脚本输出

  • 成功:脚本 exit code 0,结果在 stdout(或加 --json 时输出 JSON:successresultlogserror_message)。
  • 失败:exit code 非 0,错误信息在 stderr;根据输出提示用户(如缺 API Key 则按「API Key(仅报错时引导)」处理)。

重要约束

  • language 支持:pythonjavascriptrjava(不区分大小写)。
  • 单次执行超时 60 秒(可用 --timeout-s 指定,不超过 60)。

GitHub Repository

agentbay-ai/agentbay-skills
Path: agentbay-aio-skills
0

Related Skills

content-collections

Meta

This skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.

View skill

polymarket

Meta

This skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.

View skill

creating-opencode-plugins

Meta

This skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.

View skill

sglang

Meta

SGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.

View skill