Token导航 LogoToken导航TokenDH.com
研究检索执行命令clawhub未标认证来源可访问clear审计通过

ssh-executorssh 执行器

Agent Skill

ssh-executor 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

3,622

周安装

154

GitHub Stars

公开资料未说明

下载量

1,269
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:ssh-executor(ssh 执行器)
来源仓库:https://github.com/rickkbarbosa/ssh-executor
安装命令:
openclaw skills install ssh-executor
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 ssh-executor

简介

通过 SSH 在远程主机执行命令的安全工具。

  • 支持别名配置、tmux 会话管理与私钥认证。
  • 适用于服务器运维与自动化脚本触发场景。
  • 安装命令:openclaw skills install ssh-executor。
  • 使用前应验证连接凭证与命令执行风险。ssh-executor 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

name
ssh-executor
description
Execute commands on remote hosts over SSH using SSH aliases, ssh config, tmux sessions, and private keys. Use when the user asks to SSH into a host, inspect Linux or server state remotely, run one-off remote shell commands, reuse an SSH alias, or drive a tmux-based remote workflow. Prefer for key-based SSH access and remote diagnostics. Do not use for password storage, private-key exfiltration, silent host-key bypass, or destructive changes without explicit confirmation.
metadata

SSH Executor

Use this skill to run remote commands safely over SSH.

Quick start

  1. Gather the host alias, username, port, and key path from the user's instructions or local SSH config.
  2. Prefer SSH aliases from ~/.ssh/config instead of raw IPs when available.
  3. Default to read-only commands first.
  4. Require explicit user confirmation before destructive or state-changing commands.
  5. Use the bundled scripts/ssh-run.sh helper for execution.
  6. Return stdout, stderr, exit code, and resolved SSH metadata clearly.

Safety rules

  • Prefer key-based auth. Do not ask the user to paste passwords into chat unless they explicitly insist and understand the risk.
  • Do not reveal or copy private-key contents into chat, logs, or memory files.
  • Do not disable host-key checking silently.
  • Prefer the host's existing ssh config when possible.
  • Treat these as destructive unless the user clearly asked for them: rm, mv, chmod, chown, systemctl restart|stop|disable, reboot, shutdown, package installs/upgrades, docker compose down, schema changes, file writes, or anything using sudo.
  • For destructive work, confirm the exact command before running it, then pass --confirm-dangerous.
  • If a command can be split into inspect first and mutate later, do the inspect step first.

Workflow

1. Resolve target

Collect or infer:

  • host or SSH alias
  • user, if not already encoded in the alias
  • port, if non-default
  • key path, if needed
  • timeout
  • desired host-key policy
  • optional ssh config path

If a known SSH alias already exists, prefer it over a raw host/IP.

Only inspect ~/.ssh/config when the user already uses SSH aliases or asks you to resolve them.

To inspect available aliases from the default ssh config:

scripts/ssh-run.sh --list-aliases

To inspect aliases from a custom config file:

scripts/ssh-run.sh --list-aliases --config ~/.ssh/config

2. Decide risk level

  • Read-only: hostname, uname -a, uptime, df -h, journalctl -n 100, docker ps
  • Mutating: package management, service restarts, file edits, deletes, deploys

Read-only commands can usually run immediately. Mutating commands need explicit confirmation.

3. Run command

Use:

scripts/ssh-run.sh --host <host-or-alias> [--user <user>] [--port <port>] [--key <path>] [--timeout <seconds>] [--config <ssh-config>] [--host-key-checking accept-new|yes] -- '<command>'

If the command is mutating and the user explicitly approved it:

scripts/ssh-run.sh --host <host-or-alias> --confirm-dangerous -- '<command>'

Examples:

scripts/ssh-run.sh --host web-1 -- 'hostname && uptime'
scripts/ssh-run.sh --host 192.168.1.50 --user root --port 2222 --key ~/.ssh/id_ed25519 --timeout 15 --host-key-checking yes -- 'df -h && free -h'
scripts/ssh-run.sh --host prod-app --confirm-dangerous -- 'sudo systemctl restart myapp'

4. Report result

Summarize briefly:

  • whether the SSH connection succeeded
  • exit code
  • resolved host/user/port when useful
  • key findings from stdout/stderr
  • next safe step

Install and test

This skill lives under the workspace skills/ directory so OpenClaw can discover it in future sessions.

Suggested smoke tests for the bundled helper:

bash -n scripts/ssh-run.sh
scripts/ssh-run.sh --help
scripts/ssh-run.sh --list-aliases

Resources

  • scripts/ssh-run.sh: key-based SSH wrapper with structured JSON output, alias support, and dangerous-command confirmation
  • references/safety.md: extra guidance for safe remote execution

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

87.62%
按下载量换算1,112

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 openclaw skills install ssh-executor 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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。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