cvm-ai-doctor

cvm-ai-doctor is a skill for Claude Code, Codex from TencentCloud/Octop. It costs 60 tokens per session (5,056 once invoked), scanned A, original, MIT.

A health-diagnosis workflow for servers, computers, virtual machines, and containers on Linux, macOS, or Windows.

In plain words
What is it for?
Use it for quick health checks, deeper troubleshooting, performance advice, and—when paired with the required infrastructure tool—multi-node or cluster inspections.
Why use it?
It helps identify whether slowness, crashes, high resource use, or other system problems come from the CPU, memory, disk, network, or another component.

Skill for Claude CodeCodex

Which agent this was written for is unclear — built for openclaw. Also seen: built for openclaw.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is - Linux/macOS: bash scripts/quick_scan.sh.

Good fit Use it for quick health checks, deeper troubleshooting, performance advice, and—when paired with the required infrastructure tool—multi-node or cluster inspections.

Compare 6 skills from other repositories ↓
About the project

Octop is a self-hosted, multi-user AI assistant that runs multiple specialized agents and connects them to chat interfaces, tools, and external services. It is for individuals, families, and teams who want a locally operated assistant with shared experts and persistent capabilities. Catalogue add-ons extend its agent and assistant workflows.

TencentCloud/Octop · 1,533 stars · on GitHub · octop.cloud

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/TencentCloud/Octop
agentmods
npx agentmods add skills/tencentcloud/octop/cvm-ai-doctor

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cvm-ai-doctor

README.md
[![agentmods](https://agentmods.dev/badge/skills/tencentcloud/octop/cvm-ai-doctor/github.svg)](https://agentmods.dev/skills/tencentcloud/octop/cvm-ai-doctor)
Your own site
<a href="https://agentmods.dev/skills/tencentcloud/octop/cvm-ai-doctor"><img src="https://agentmods.dev/badge/skills/tencentcloud/octop/cvm-ai-doctor/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cvm-ai-doctor

Your own site · 80×15
<a href="https://agentmods.dev/skills/tencentcloud/octop/cvm-ai-doctor"><img src="https://agentmods.dev/badge/skills/tencentcloud/octop/cvm-ai-doctor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 60 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,056 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00060 $0.05056
Opus 5 $0.00030 $0.02528
Sonnet 5 $0.00012 $0.01011
Haiku 4.5 $0.00006 $0.00506

Measured 5d ago against content hash 69eb113e6787, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

cvm-ai-doctor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

src/octop/infra/agents/experts/library/cvm-ai-doctor/skills/cvm-ai-doctor/SKILL.md · 470 lines

How it starts

The opening of the file, as written. The whole thing — 470 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CVM AI Doctor 技能

用途: 智能诊断系统各类性能问题和使用问题,采用快速检查(10秒分诊)→ 深度分析(20-60秒)工作流。

支持系统: 本地服务器、个人计算机(Mac/Windows/Linux)、虚拟机、容器。


🎯 何时使用此技能

核心能力:

  • 系统性能诊断(CPU、内存、磁盘、网络)
  • 健康检查(快速扫描、综合分析)
  • 系统故障排查(崩溃、错误、慢、卡)
  • 性能优化建议
  • 集群管理:多节点健康巡检、评分、关联分析、风险门控修复(需配合 tencentcloud-infra 技能)

触发关键词: 详见"Tier 1 核心场景"中的完整关键词列表

注意: 如果用户提到"LightClaw 自检"或"平台体检"或"安全检查",请使用 health-check 技能。

注意: 如果用户提到"集群"、"所有节点"、"所有CVM",使用集群管理场景,需同时加载 tencentcloud-infra 技能。

🌐 远程服务器

  • 用户提到 IP 地址远程主机 → 结合 remote-connect 技能使用(详见"远程诊断"章节)

🚀 快速/深度诊断工作流

两阶段诊断方法:

快速模式 (3-10秒):
  - 快速扫描所有组件
  - 执行策略:
      优先: 执行快速扫描脚本 (如果存在) → ~3秒
        - Windows: pwsh scripts/quick_scan.ps1
        - Linux/macOS: bash scripts/quick_scan.sh
      回退: 读取 resource-saturation-quick.md → 运行单独命令 → ~10秒
  - 返回: 每个组件的 OK / WARNING / CRITICAL 状态
  
深度模式 (20-60秒, 按需触发):
  - 单组件: 只分析标记组件
  - 多组件: 使用组合关联分析
  - 根本原因分析 + 建议

决策规则:
  - 通用问题 ("慢", "卡") → 先快速扫描,再对标记组件深度分析
  - 特定组件 ("CPU高") → 跳过快速,直接深度分析
  - 多个异常 (2+ 个标记) → 使用深度组合分析
  - 健康检查 → 仅快速扫描,全部 OK 则停止

🔍 场景匹配

Tier 1: 核心场景 (80% 覆盖率)

优先匹配以下场景:

系统慢/卡 (System Slow):
  keywords: [慢, 卡, slow, lag, 响应慢, sluggish, 性能问题, performance issue, 性能故障, performance problem]
  workflow: 快速扫描 → 对标记组件深度分析
  fixes: [commands/debugging/diagnose-slowdown.md, commands/system-health/review-startup-services.md]

CPU 高 (High CPU):
  keywords: [CPU, cpu高, cpu 100%, CPU满]
  workflow: 直接使用 resource-saturation-deep-cpu.md
  fixes: [commands/hardware/hardware-profilers/by-component/profile-cpu.md]

内存不足 (Low Memory):
  keywords: [内存, memory, OOM, swap, 内存不足]
  workflow: 直接使用 resource-saturation-deep-memory.md
  fixes: [commands/hardware/hardware-profilers/by-component/profile-ram.md]

磁盘问题 (Disk Issues):
  keywords: [磁盘, disk, I/O, io慢, 磁盘满, 磁盘空间, 盘满, C盘, D盘, disk full, space, 存储]
  workflow: 直接使用 resource-saturation-deep-disk.md
  fixes: [commands/optimisation/large-files.md, commands/storage/health-checks/check-drive-health.md]

网络问题 (Network Issues):
  keywords: [网络, network, 网卡, 丢包]
  workflow: 直接使用 resource-saturation-deep-network.md
  fixes: [commands/network/lan/diagnose-lan-connectivity.md]

健康检查 (Health Check):
  keywords: [检查, 健康, health, check, 诊断, 体检, 健康度, inspection, 专业检查, 专家检查, 快速检查, 深入检查, 深度检查, 全面检查]
  workflow: |
    一般健康检查 → 运行快速扫描脚本 (3秒):
      全部 OK 则停止。
    综合健康检查 (用户说"综合检查" / "全面体检" / "系统全检" / "深入检查" / "深度检查" / "专业检查" / "专家检查"):
      按顺序运行 3 个快速模块:
        1. resource-saturation-quick.md (CPU/内存/磁盘/网络)
        2. system-logs-quick.md (内核/OOM/文件系统/认证/服务错误)
        3. disk-smart-quick.md (磁盘 SMART 健康度)
      对发现的任何 WARNING/CRITICAL 组件进行深度分析。
  fixes: [commands/system-health/system-health-checkup.md]

系统崩溃/错误 (System Crash):
  keywords: [崩溃, crash, 错误, error, 重启, reboot, 系统救援, 系统急救, emergency, rescue, 紧急修复, 故障修复]
  workflow: system-logs-quick.md → 对错误类别深度分析
  fixes: [commands/debugging/diagnose-crash.md]

性能优化 (Performance Optimization):
  keywords: [性能优化, 性能提升, 提升性能, performance optimization, performance tuning, optimize, tuning, 调优, 优化建议]
  workflow: 快速扫描 → 识别瓶颈 → 深度分析 → 优化建议
  fixes: [commands/hardware/hardware-profilers/hardware-profile.md, commands/optimisation/large-files.md]

磁盘健康 (Disk Health):
  keywords: [磁盘健康, SMART, 坏道, 硬盘故障]
  workflow: disk-smart-quick.md → 如果 FAILED 则 disk-smart-deep.md
  fixes: [commands/storage/health-checks/check-drive-health.md]

硬件问题 (Hardware Issues):
  keywords: [硬件, hardware, 温度, RAID, ECC, 风扇]
  workflow: hardware-health-quick.md → hardware-health-deep.md
  fixes: [commands/hardware/hardware-profilers/hardware-profile.md]

时间问题 (Time Issues):
  keywords: [时间, NTP, 时钟, 时间不对, Kerberos]
  workflow: time-sync.md 快速 → 如未同步则深度分析
  fixes: []  # time-sync.md 本身包含修复步骤

SSH 登录问题 (SSH Login Issues):
  keywords: [SSH, ssh登录, 无法登录, 登录失败, can't login, login failed, connection refused, 连不上]
  workflow: |
    使用 cvm-self-diagnosis-repair.md → 诊断 SSH 服务、防火墙、磁盘、内存
    需用户确认后修复: 重启服务、修复配置、开放防火墙
  fixes: [references/cvm-self-diagnosis-repair.md]

诊断统计 (Diagnosis Stats):
  keywords: [统计, 诊断统计, stats, statistics, 诊断报告, 历史诊断, 诊断历史, 诊断趋势, diagnosis report, 问题率, 诊断次数]
  workflow: |
    执行: bash scripts/analyze_stats.sh
    支持的参数:
      --today     仅今天的数据
      --last Nd   最近 N 天 (如 --last 7d)
      --json      输出 JSON (供程序消费)
    如果用户问特定时间段 → 加对应参数
    如果用户问总览/概况 → 不加参数(全量)
  fixes: []

集群管理 (Cluster Management):
  keywords: [集群, cluster, 所有CVM, 所有节点, fleet, 批量检查, 多台服务器, 巡检, 节点, 健康巡检, 集群健康, cvm集群, 节点状态]
  workflow: |
    前置:本场景所有 tccli 命令通过 tencentcloud-infra 技能执行,
          必须确认该技能在当前会话已加载,且 OAuth 凭据有效。

    路由判断:

    1. 无集群配置(MEMORY.md 无 cluster_config 段落)→ 首次配置引导
       读 references/cluster-discovery.md,向用户展示以下引导词:
       ─────────────────────────────────────────────────────────
       我需要知道要管理哪些 CVM 实例。请选择配置方式:

       1. 标签过滤(推荐):按服务标签批量选择,例如 Service=order-api
       2. 指定实例 ID:直接输入 ins-xxx,ins-yyy,ins-zzz
       3. 全量扫描:当前区域所有实例(⚠️ 大账户请谨慎)

       请告诉我:
       - 选哪种方式?
       - 实例所在区域(如 ap-guangzhou)?
       - 对应的标签键值 或 实例 ID 列表?
       ─────────────────────────────────────────────────────────
       收到回复后,通过 tencentcloud-infra 执行 DescribeInstances 并展示结果让用户确认,
       确认后写入 MEMORY.md cluster_config 段落。

    2. 有配置,用户要健康检查:
       → 读 references/cluster-quick-check.md(3 层快照,约 30 秒)
       → 读 references/cluster-health-score.md(生成评分报告)

    3. 发现异常,用户要深度分析:
       → 读 references/cluster-deep-analysis.md(按异常类型选 TAT 命令)

    4. 用户要修复:
       → 读 references/cluster-remediation.md(风险门控,串行执行)

    协作协议见 references/skill-collaboration-tencentcloud.md
  fixes: [references/cluster-remediation.md]
  risk_level: 🟡~🔴(依据具体操作,读 cluster-remediation.md 风险等级)

Read the full file on GitHub · 470 lines

Files

What ships with it

60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago Changed · +12 lines 69eb113e6787
  2. 12d ago First seen · 458 lines · 60 tokens per session scan A d5d821d7c99f

Subscribe to this mod's changes

cvm-ai-doctor is a skill published in the GitHub repository TencentCloud/Octop (1,533 stars, last pushed today), licensed MIT. It adds 60 tokens to every session and 5,056 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

rubber-duck

Adversarial "rubber duck" review that turns explaining-out-loud into a hallucination check. The main session is the PRESENTER (it did the work — a design doc, investigation, or analysis — and holds the real reasoning) and reconstructs the topic to a LISTENER — a spawned subagent pinned to a DIFFERENT-vendor model that…

kirodotdev/KiroCrew · 161 tokens

code-search

Search a codebase efficiently with ripgrep regular expressions, file globs, and git history search. Use to locate symbols, usages, and definitions instead of reading whole files.

agentscope-ai/agentscope-java · 38 tokens

agent-identity

Verify an agent's live address and billing account before sharing a chat link, diagnosing a deploy mismatch, or deciding why two hosts share an identity.

openonion/connectonion · 33 tokens

ai-discover

Parallel discovery of performance hotspots (perf track) and failure surfaces (bug track) for the auto-improvement loop. Fans out one subagent per hot-path area or failure surface; each returns ONE concrete, behavior-preserving fix candidate (perf) or a reproducing test plus fix (bug). Discovery only — no code changes…

kirodotdev/KiroCrew · 75 tokens

debugging-websocket-issues

Use when seeing WebSocket errors like "Invalid frame header", "RSV1 must be clear", or "WSERRUNEXPECTEDRSV1" - covers multiple WebSocketServer conflicts, compression issues, and raw frame debugging techniques.

AgentWorkforce/relay · 54 tokens

gini-bug-report

File a locally-captured, already-redacted Gini crash report as a GitHub issue, with the user's consent. Reads the pending crash queue and delegates the actual filing to the github-issues skill.

Open-Curiosity/gini-agent · 47 tokens