← 返回 Skills 市场
Python Executor
作者
Ömer Karışman
· GitHub ↗
· v0.1.5
9981
总下载
3
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install python-executor
功能描述
Execute Python code in a safe sandboxed environment via [inference.sh](https://inference.sh). Pre-installed: NumPy, Pandas, Matplotlib, requests, BeautifulSo...
安全使用建议
Install only if you trust inference.sh and intend to let an agent run Python code through that provider. Prefer manual install with checksum verification over piping an installer into a shell, review code before execution, and avoid sending secrets, private datasets, credentials, or confidential generated files unless you explicitly intend that data to be processed or returned by the service.
功能分析
Type: OpenClaw Skill
Name: python-executor
Version: 0.1.5
The skill allows the AI agent to execute arbitrary Python code within a sandboxed environment via `infsh app run`. The `allowed-tools: Bash(infsh *)` permission is broad, granting the agent the ability to execute any `infsh` command. While the `SKILL.md` claims a 'safe sandboxed environment' and does not contain explicit malicious instructions for the agent, the inherent capability to execute arbitrary code (even if sandboxed) and the broad `infsh` permissions present a significant risk if the agent were to be compromised via prompt injection, potentially leading to the execution of malicious Python code or other `infsh` commands.
能力评估
Purpose & Capability
The skill is explicitly for executing Python through inference.sh, including web scraping, API calls, media processing, automation, and returning generated files; these are powerful capabilities but coherent with the stated purpose.
Instruction Scope
The `Bash(infsh *)` permission and broad trigger terms could invoke a powerful tool in more situations than ideal, but the authority is limited to the advertised infsh workflow and no unrelated commands or concealed instructions were found.
Install Mechanism
The quick start recommends `curl ... | sh` and `infsh login`, which is a supply-chain and account-session consideration; the artifact discloses this and also mentions checksum/manual verification, no elevated permissions, and no background processes.
Credentials
Network-capable Python libraries, browser automation, remote execution, and automatic output return are proportionate for this kind of executor, but users should treat submitted code, inputs, network calls, and returned files as potentially crossing a provider boundary.
Persistence & Privilege
Only a markdown skill artifact was present; no bundled executables, hidden persistence, privilege escalation, destructive actions, local credential harvesting, or background workers were identified.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install python-executor - 安装完成后，直接呼叫该 Skill 的名称或使用
/python-executor触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v0.1.5
- Added detailed documentation in SKILL.md, including usage examples, input schema, and pre-installed library list.
- Clarified environment specs, supported triggers, and output file handling.
- Listed practical use cases for web scraping, data analysis, image/video/3D processing, API calls, PDF generation, and automation.
- Provided quick start instructions and links to related skills and official documentation.
v0.1.0
Initial release – execute Python code in a sandboxed environment with 100+ libraries.
- Run Python scripts safely using the inference.sh platform.
- Pre-installed support for libraries such as NumPy, Pandas, Matplotlib, requests, BeautifulSoup, Selenium, Playwright, MoviePy, Pillow, OpenCV, trimesh, and more.
- Supports data processing, web scraping, image/video editing, 3D models, PDF generation, API calls, and automation.
- Files saved to outputs/ are automatically returned.
- Choose between 8GB (default) or 16GB (high_memory) RAM variants.
元数据
常见问题
Python Executor 是什么？
Execute Python code in a safe sandboxed environment via [inference.sh](https://inference.sh). Pre-installed: NumPy, Pandas, Matplotlib, requests, BeautifulSo... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 9981 次。
如何安装 Python Executor？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install python-executor」即可一键安装，无需额外配置。
Python Executor 是免费的吗？
是的，Python Executor 完全免费（开源免费），可自由下载、安装和使用。
Python Executor 支持哪些平台？
Python Executor 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Python Executor？
由 Ömer Karışman（@okaris）开发并维护，当前版本 v0.1.5。
推荐 Skills