Menu
Execute shell commands and scripts on the user's local machine.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Display interactive HTML content, visualizations, dashboards, or games on connected OpenClaw nodes (Mac, iOS, Android). Supports live reloading, remote navigation, JS execution, and canvas snapshots for development, testing, and presentation scenarios.
Use the ClawHub CLI to search, install, update, and publish agent skills from clawhub.com. Use when you need to fetch new skills on the fly, sync installed skills to latest or a specific version, or publish new/updated skill folders with the npm-installed clawhub CLI.
多模态 AI 图片理解工具。将图片发送给支持 vision 的 LLM 模型，获取图片的语义理解和详细描述。可分析截图、设计稿、图表、报错信息等。
Unified memory search and management system for retrieving, storing, and organizing information across daily logs, persistent memory, SOPs, and execution traces. Use this skill to recall past events, manage knowledge bases, update important notes, or filter historical data by tags and time ranges.
智能 OCR 文字提取工具。从图片中识别文字，支持中英文混排、图像预处理（灰度化+二值化提升识别率）、表格结构还原为 Markdown 格式。
Batch-generate images via OpenAI Images API. Random prompt sampler + `index.html` gallery. Ideal for rapid prototyping, generating placeholder assets, testing AI art variations across models (GPT-Image, DALL-E 3/2), and building local thumbnail galleries with customizable resolution, quality, and output formats.
| name | code-runner |
| description | Execute shell commands and scripts on the user's local machine. |
| version | 1.0.0 |
| author | DunCrew |
| tags | ["shell","execution","system-ops","windows"] |
| keywords | ["run command","execute script","terminal","cmd","system diagnostics","package management"] |
| dangerLevel | high |
| inputs | [{"name":"command","type":"string","description":"The shell command or script path to execute on the local machine.","required":true}] |
| metadata | {"openclaw":{"emoji":"⚡","primaryEnv":"shell"}} |
Execute shell commands and scripts on the user's local machine.
{"thought": "...", "tool": "runCmd", "args": {"command": "..."}}
User: "查看当前目录有哪些文件"
{"thought": "用户想查看当前目录的文件列表，使用 dir 命令（Windows 系统）", "tool": "runCmd", "args": {"command": "dir /b"}}
User: "帮我创建一个 Python 虚拟环境并安装 requests 库" Step 1:
{"thought": "需要先创建虚拟环境，使用 python -m venv 命令", "tool": "runCmd", "args": {"command": "python -m venv myenv"}}
Step 2 (after success):
{"thought": "虚拟环境创建成功，现在激活并安装 requests。Windows 下用 myenv\\Scripts\\pip", "tool": "runCmd", "args": {"command": "myenv\\Scripts\\pip install requests"}}
Step 3 (verify):
{"thought": "安装完成，验证一下 requests 是否正确安装", "tool": "runCmd", "args": {"command": "myenv\\Scripts\\pip list | findstr requests"}}
User: "看看端口 3001 被谁占用了"
{"thought": "需要查看端口占用情况，使用 netstat 命令过滤 3001 端口", "tool": "runCmd", "args": {"command": "netstat -ano | findstr 3001"}}
If previous command failed with "python not found":
{"thought": "python 命令未找到，可能系统用的是 python3 或完整路径。先检查 Python 安装位置。", "tool": "runCmd", "args": {"command": "where python"}}