Ssh Remote Sanitized

SSH 远程服务器管理工具。支持多服务器连接管理、命令执行、文件传输、系统监控、服务管理、日志分析、安全检查等功能，适用于 Linux/Unix 服务器运维。

This skill implements an SSH management tool that will: (1) read/write server configs under ~/.openclaw/skills/ssh-remote/configs/servers.json, (2) optionally read SSH private keys from paths you supply (e.g. ~/.ssh/id_rsa) or environment variables, and (3) execute arbitrary commands on your remote servers and run package installation via npm. The registry metadata did not list any required environment variables even though the code uses several — that's an inconsistency you should be aware of. Before installing: inspect configs/servers.json (don’t store private keys in the skill directory), review package.json and run npm install and npm audit in a safe environment, consider using a dedicated least-privilege SSH account (not root) for this skill, and avoid granting it access to your primary private keys unless you trust the source. If you need higher assurance, review the src/*.js files yourself or run the skill in an isolated container/VM.

Type: OpenClaw Skill Name: ssh-remote-sanitized Version: 1.0.0 The skill bundle provides a comprehensive SSH management tool, but it contains multiple shell injection vulnerabilities across several files (e.g., src/log.js, src/service.js, src/upload.js, and src/security.js). User-provided inputs like file paths, service names, and search patterns are directly concatenated into command strings executed via SSH (e.g., in `searchLog` and `getServiceStatus`) without any sanitization or escaping. While these appear to be unintentional security flaws rather than intentional malice, they represent a significant risk as they allow for arbitrary command execution on the remote server if the AI agent is manipulated.