用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/taracodlabs/aiden --skill explainshell命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | explainshell |
| description | Explain shell commands in plain English (explainshell.com) |
| category | developer |
| version | 1.0.0 |
| license | Apache-2.0 |
| origin | aiden |
| tags | shell, bash, linux, commands, explainer, learning, devops, documentation, scripting |
Paste any shell command or one-liner and get a plain-English breakdown of every flag, argument, and pipe stage. Powered by explainshell.com, which maps commands to their man-page descriptions.
No API key required. Works immediately for any user.
# The simplest approach — open the explanation in the browser
$command = "tar -xzf archive.tar.gz -C /tmp"
$encoded = [Uri]::EscapeDataString($command)
Start-Process "https://explainshell.com/explain?cmd=$encoded"
Write-Host "Opened: https://explainshell.com/explain?cmd=$encoded"
$command = "find . -type f -name '*.log' -mtime +7 -exec rm {} \;"
$encoded = [Uri]::EscapeDataString($command)
$url = "https://explainshell.com/explain?cmd=$encoded"
$html = Invoke-WebRequest -Uri $url -UseBasicParsing
# Extract helptext spans (plain text explanations)
$pattern = '<span class="helptext">(.*?)</span>'
$matches = [regex]::Matches($html.Content, $pattern, 'Singleline')
Write-Host "Command: $command"
Write-Host ""
$matches | ForEach-Object {
$text = $_.Groups[1].Value -replace '<[^>]+>', '' -replace '&', '&' -replace '<', '<' -replace '>', '>'
if ($text.Trim()) { Write-Host " • $($text.Trim())" }
}
Write-Host ""
Write-Host "Full explanation: $url"
$command = "ps aux | awk '{print \$2, \$11}' | sort -k2 | uniq -c | sort -rn | head -10"
$encoded = [Uri]::EscapeDataString($command)
Write-Host "ExplainShell link:"
Write-Host " https://explainshell.com/explain?cmd=$encoded"
"Explain: tar -xzf file.tar.gz"
→ tar: archive utility; -x: extract; -z: filter through gzip; -f: use archive file.
"What does find . -type f -exec rm {} \; do?"
→ find: search; .: starting directory; -type f: only files; -exec rm {} ;: run rm on each result.
"Break down: awk '{print $2}' | sort | uniq -c"
→ awk: text processor prints second field; sort: lexicographic sort; uniq -c: count consecutive duplicates.
"What is 2>&1 in a command?" → Redirects stderr (fd 2) to the same destination as stdout (fd 1) — merges error output.
awk programs) parse partially