🩺 ops-doctor

One command. Full server checkup. No agents. No config. No dependencies.

curl -fsSL https://raw.githubusercontent.com/tomlen045/ops-doctor/main/ops-doctor.sh | bash

Paste that on any Linux server and get an instant health report: load, memory, disk, inodes, zombies, failed services, OOM history, pending reboots, SSH hardening — with a 0-100 health score and concrete fix suggestions. Not a wall of raw numbers: every finding tells you what to do about it.

Why

Servers don't fail dramatically — they rot quietly. Disk creeps to 90%, a service dies three weeks ago, OOM started killing things last Tuesday. ops-doctor catches all of that in the 5 seconds before your boss asks why the site is down.

The author ran it on his own Mac before publishing. It found real problems. That's the demo.

Sample output

== Disk ==
  [CRIT] disk /data is 93% full — writes will fail soon
  [WARN] disk / is 81% full
  [ OK ] disk usage scan done (alert threshold 80%)

== OOM history ==
  [ OK ] no OOM kills in the last 7 days

==== HEALTH SCORE: 0/100 🚨 UNHEALTHY ====
  critical: 10 · warnings: 1

Features

• 15+ checks: load, memory/swap, disk space and inodes, zombies, failed systemd units, OOM kills, reboot-required, pending updates, SSH hardening (root login / password auth)
• Health score 0–100 with exit code (0 healthy, 1 critical) — drop it straight into cron or CI
• Markdown report: bash ops-doctor.sh --md report.md — paste into tickets or wikis
• Zero dependencies: bash + coreutils. No agent to install, nothing left behind
• Private by design: runs locally, sends nothing anywhere

Usage

bash ops-doctor.sh                 # colored report + health score
bash ops-doctor.sh --md report.md  # also write a markdown report
echo $?                            # 0 = healthy, 1 = critical findings

Cron example — weekly checkup emailed to you

0 9 * * 1 bash /opt/ops-doctor.sh --md /tmp/report.md && mail -s "weekly server checkup" you@company.com < /tmp/report.md

Compatibility

Linux (primary — designed for servers) · macOS (partial, core checks only)

中文说明：一条命令给 Linux 服务器做全套体检，输出健康分和修复建议。纯本地运行，零依赖，不上传任何数据。

License

MIT

---

🫰 关注「小薅薅」· 每天发现一个宝藏工具

小薅薅 — 年轻人的赛博工具箱。每天为你发现一个好玩、实用、开源的小工具，帮你节省 1 小时探索时间。

📱 微信搜索「小薅薅」 或扫描下方二维码关注

后台回复关键词获取：

• 回复 「工具」 → 获取全部工具合集（离线可用）
• 回复 「运维」 → 获取运维/安全资源包
• 回复 「加群」 → 加入工具交流群

🎯 更多作品	描述	Stars
ops-skills	10个AI运维技能包，让Claude Code变成SRE专家	⭐ 新项目
shellmbti	你的终端历史暴露了你是谁——Shell MBTI人格测试	⭐ 病毒传播
naicha-mbti	8道题测出你的奶茶人格，生成分享卡片	🧋 爆款
fafa-generator	发疯文学生成器——一键生成发疯文案+卡片	🔥 热门
life-progress	人生进度条——把你的时间摆在眼前	⏳ 走心

觉得有用？给个 ⭐ 让更多人看到 → GitHub | Gitee