系统健康检查

SKILL运维和基础设施
运维和基础设施 未标注权限 许可证不明确
安装这个技能

【系统健康检查】(System Health Check)。当用户想检查电脑健康状况、看 CPU 负载、内存占用、存储空间、电池健康度,或提到『检查电脑、系统体检、电脑卡不卡、磁盘满了、内存不够、电池健康、CPU 占用高、清理空间』时使用。核心工作法:运行内置采集脚本(纯标准库、跨平台 macOS/Windows/Linux/HarmonyOS、离线可用),一次采集 CPU 使用率与负载均值、内存占用、存储空间(含各卷明细)、电池健康度(含循环次数与电源状态),自动评级(良好/注意/警告)并计算综合健康分,渲染为自包含单文件 HTML 报告,随后向用户汇报关键数据与需要关注的风险点。

一句话选型结论

按资料字段整理

系统健康检查是运维和基础设施类的 Agent Skill,安装方式未标注,权限级别未标注,许可证不明确,有可直接复制的安装命令。技能本质是给模型读的说明书,效果取决于模型与客户端;装完先用一个小任务验证,再放进正式流程。

分类运维和基础设施 安装方式未标注 权限级别未标注权限 许可证许可证不明确
累计安装 未标注 本周 未标注
GitHub 星标 未标注 仓库 未标注
下载量 未标注 技能市场下载数
权限级别 未标注权限 安装前请确认风险范围
许可证 许可证不明确 未说明

安装

未标注
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
skillhub install xh-ai/system-health-check
  1. 1
    确认客户端支持

    确认你的客户端支持 Skills(Claude、Codex、Cursor 等),并升级到较新的版本。

  2. 2
    拿到技能文件

    执行上面的命令安装;也可以把仓库里的 SKILL.md 手动放进客户端的 skills 目录。

  3. 3
    刷新技能列表

    重新打开客户端,或刷新技能列表,让新放进来的技能被加载。

  4. 4
    验证是否生效

    问一句「你有哪些技能」,确认列表里出现它,再拿一个小任务试跑。

技能档案

运维和基础设施
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方xh-ai仓库 未标注 · 版本 未标注
宿主客户端未标注技能可被这些客户端加载
安装方式未标注安装包 enterprise
权限级别未标注权限安装前请确认风险范围
许可证未说明许可证不明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验未标注上游最后核验时间 · 数据更新 1791424001
正文体量14.4 KBSKILL.md 原文大小,越大说明说明书写得越细

典型使用场景

3 类
按分类整理的常见用法,具体以 SKILL.md 为准
查服务与日志

按现象定位服务状态与日志线索。

写部署与巡检脚本

把重复的部署、巡检写成脚本。

故障排查

按步骤排查网络、依赖与配置。

选择建议

装之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 有可直接复制的安装命令,不用自己翻文档找入口。

先想清楚的地方

  • 没有标注权限级别,实际能做什么要装好后确认。
  • 许可证不明确,商用或二次分发前先到仓库确认条款。
  • 没有做站内核验,内容以作者仓库为准。
  • 技能只是说明书,实际效果取决于模型与客户端,装完先用小任务试一遍。

更适合

  • 需要在运维和基础设施场景里给模型补一套做法的人。
  • 愿意先读一遍 SKILL.md、确认它会让模型做什么的人。

不太适合

  • 对许可证要求很严、必须明确可商用的人。
  • 只想装完就自动变强、不打算看说明书的人。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
技能权限级别安装方式累计安装宿主客户端安装命令
系统健康检查当前 未标注权限 未标注 未标注 未标注 有
Azure 部署 调用外部服务 GitHub 仓库 664.2 万 Codex、Claude、Cursor 有
Azure diagnostics 问题管理 调用外部服务 GitHub 仓库 657.6 万 Codex、Claude、Cursor 有
Azure aigateway 部署 调用外部服务 GitHub 仓库 655.3 万 Codex、Claude、Cursor 有
Azure AI 搜索 调用外部服务 GitHub 仓库 651.4 万 Codex、Claude、Cursor 有
Azure Cloud migrate 部署 调用外部服务 GitHub 仓库 650.7 万 Codex、Claude、Cursor 有

SKILL.md 全文

14.4 KB
模型实际读到的说明书原文,来自技能仓库

System Health Check

This skill validates the entire OpenClaw Greek Accounting system in a single command. It checks skill file integrity, directory structure, file permissions, required dependencies, backup freshness, encryption status, and process lock health. Designed to run daily via cron or manually before critical operations.

Setup

export OPENCLAW_DATA_DIR="/data"
which jq openssl || sudo apt install jq openssl

Read-only validation skill. Checks file presence, directory structure, permissions, and backup freshness. Never modifies any data.

Core Philosophy

  • Fast & Non-Destructive: Read-only checks — never modifies any data
  • Comprehensive: Covers every layer from skill files to encryption status
  • Actionable Output: Every failure includes a specific remediation command
  • Cron-Friendly: Exit code 0 for all-pass, exit code 1 for any failure
  • English Output: Plain English report suitable for accounting assistants and system admins

OpenClaw Commands

Full Health Check

# Run all checks
openclaw health-check --all

# Run all checks with verbose output
openclaw health-check --all --verbose

# Run all checks and write report to /data/reports/system/
openclaw health-check --all --save-report

Individual Check Categories

# Check skill files only
openclaw health-check --skills

# Check directory structure against canonical data map
openclaw health-check --directories

# Check file permissions on sensitive directories
openclaw health-check --permissions

# Check required binaries and dependencies
openclaw health-check --dependencies

# Check backup freshness and integrity
openclaw health-check --backups

# Check for stale process locks
openclaw health-check --locks

# Check encryption status on sensitive directories
openclaw health-check --encryption

# Check data integrity hashes
openclaw health-check --integrity

Scheduled Checks

# Quick check — skills + directories + locks (for hourly cron)
openclaw health-check --quick

# Standard check — everything except full integrity hash verification (for daily cron)
openclaw health-check --standard

# Deep check — all checks including full hash verification (for weekly cron)
openclaw health-check --deep

Check Specifications

1. Skill File Validation

Verifies all 19 operational skills and the canonical data map are present and well-formed.

Skill_File_Checks:
  presence:
    description: "Every skill folder contains a SKILL.md file"
    expected_count: 20
    check: "ls skills/*/SKILL.md"

  frontmatter:
    description: "Every SKILL.md has valid YAML frontmatter with required fields"
    required_fields: ["name", "description", "version", "author", "tags", "metadata"]
    check: "Parse frontmatter block between --- delimiters"

  evals:
    description: "Every operational skill has an EVALS.json file"
    expected_count: 19
    excluded: ["canonical-data-map"]
    check: "ls skills/*/EVALS.json, validate JSON syntax"

  evals_content:
    description: "Each EVALS.json contains at least 5 test cases"
    minimum_cases: 5
    check: "Parse JSON, count array length"

  skill_inventory:
    - "canonical-data-map"
    - "accounting-workflows"
    - "greek-compliance-aade"
    - "cli-deadline-monitor"
    - "greek-email-processor"
    - "greek-individual-taxes"
    - "openclaw-greek-accounting-meta"
    - "aade-api-monitor"
    - "greek-banking-integration"
    - "greek-document-ocr"
    - "efka-api-integration"
    - "dashboard-greek-accounting"
    - "client-data-management"
    - "user-authentication-system"
    - "conversational-ai-assistant"
    - "greek-financial-statements"
    - "client-communication-engine"
    - "system-integrity-and-backup"
    - "analytics-and-advisory-intelligence"
    - "memory-feedback"

2. Directory Structure Validation

Checks all canonical directories exist under /data/ as specified in the canonical data map (Skill 00).

Directory_Checks:
  top_level:
    required:
      - "/data/incoming/"
      - "/data/processing/"
      - "/data/clients/"
      - "/data/compliance/"
      - "/data/banking/"
      - "/data/ocr/"
      - "/data/efka/"
      - "/data/reports/"
      - "/data/exports/"
      - "/data/imports/"
      - "/data/dashboard/"
      - "/data/auth/"
      - "/data/backups/"
      - "/data/gdpr-exports/"
      - "/data/memory/"
      - "/data/system/"

  subdirectories:
    incoming:
      - "/data/incoming/invoices/"
      - "/data/incoming/receipts/"
      - "/data/incoming/statements/"
      - "/data/incoming/government/"
      - "/data/incoming/payroll/"
      - "/data/incoming/tax-documents/"
      - "/data/incoming/contracts/"
      - "/data/incoming/other/"
    processing:
      - "/data/processing/ocr/"
      - "/data/processing/classification/"
      - "/data/processing/reconciliation/"
      - "/data/processing/compliance/"
    system:
      - "/data/system/logs/"
      - "/data/system/logs/audit/"
      - "/data/system/process-locks/"
      - "/data/system/integrity/"

  unexpected_directories:
    description: "Flag any top-level directory under /data/ not in the canonical map"
    action: "WARN — may indicate a skill creating non-canonical paths"

3. File Permission Checks

Validates OS-level file permissions on sensitive directories.

Permission_Checks:
  restricted_directories:
    - path: "/data/auth/"
      expected_mode: "700"
      description: "Auth directory must be restricted to service user"
    - path: "/data/auth/users/*/credentials.json"
      expected_mode: "600"
      description: "Credential files must not be world-readable"

  sensitive_directories:
    - path: "/data/clients/"
      expected_mode: "700"
      description: "Client data directory should be restricted"
    - path: "/data/backups/"
      expected_mode: "700"
      description: "Backup directory should be restricted"

  note: "Permission checks may report SKIP on systems where the OpenClaw agent runs as root or in a container without OS-level permission enforcement."

4. Dependency Checks

Verifies required binaries and environment variables are available.

Dependency_Checks:
  required_binaries:
    - name: "jq"
      check: "which jq"
      used_by: "All skills — JSON processing"
    - name: "curl"
      check: "which curl"
      used_by: "AADE monitor, email processor, EFKA, memory-feedback"
    - name: "openssl"
      check: "which openssl"
      used_by: "System integrity, backup encryption"

  required_env_vars:
    - name: "OPENCLAW_DATA_DIR"
      description: "Root data directory path"
      default: "/data/"

  optional_env_vars:
    - name: "OPENCLAW_ENCRYPTION_KEY"
      description: "Master encryption key for data-at-rest"
      warn_if_missing: true
    - name: "GITHUB_TOKEN"
      description: "GitHub PAT for memory-feedback PR workflow"
      warn_if_missing: true
    - name: "SMTP_HOST"
      description: "Mail server for client communications"
      warn_if_missing: true

5. Backup Freshness

Checks that backups are current per the schedule defined in Skill 17.

Backup_Checks:
  weekly_full:
    description: "Full backup should exist from within the last 7 days"
    location: "/data/backups/"
    pattern: "full_*.tar.enc"
    max_age_days: 7
    severity: "HIGH if missing"

  daily_incremental:
    description: "Incremental backup should exist from within the last 24 hours"
    location: "/data/backups/"
    pattern: "incremental_*.tar.enc"
    max_age_hours: 26
    severity: "MEDIUM if missing (allows 2-hour grace period)"

  verification:
    description: "Last backup verification should be within 7 days"
    check: "Look for verification report in /data/reports/system/"
    severity: "MEDIUM if stale"

6. Process Lock Health

Detects stale process locks that may indicate crashed operations.

Lock_Checks:
  location: "/data/system/process-locks/"
  stale_threshold_minutes: 30
  check_method:
    - "List all .lock files"
    - "Read timestamp from each lock file"
    - "Flag locks older than threshold"
    - "Check if process ID in lock file is still running"
  actions:
    stale_found: "WARN — list stale locks with age and owning process"
    suggestion: "Run: openclaw integrity clear-stale-locks --confirm"
    active_found: "INFO — list active locks (normal operation)"

7. Encryption Status

Verifies encryption configuration on directories that require it per the canonical data map.

Encryption_Checks:
  mandatory_encrypted:
    - "/data/auth/"
    - "/data/clients/"
    - "/data/compliance/"
    - "/data/efka/"
    - "/data/gdpr-exports/"
  check_methods:
    luks: "Check if volume is LUKS-encrypted via lsblk or cryptsetup status"
    fscrypt: "Check fscrypt policy on directory"
    env_key: "Verify OPENCLAW_ENCRYPTION_KEY environment variable is set"
  note: "In development/testing environments, encryption may not be configured. Report as WARN, not FAIL."

8. Data Integrity

Validates file hashes against the integrity registry maintained by Skill 17.

Integrity_Checks:
  registry_location: "/data/system/integrity/"
  check_method:
    - "Read hash registry for canonical files"
    - "Recompute SHA-256 hashes for each file"
    - "Compare against stored hashes"
    - "Flag any mismatches"
  scope:
    quick: "Check 10% random sample of files"
    standard: "Check all files modified in last 7 days"
    deep: "Check all files in registry"
  severity:
    mismatch: "HIGH — file has been modified outside normal skill operations"
    missing_from_registry: "MEDIUM — file exists but has no hash recorded"
    missing_file: "HIGH — registry references a file that no longer exists"

Output Format

Console Output

╔══════════════════════════════════════════════════╗
║   OpenClaw Greek Accounting — Health Report      ║
╠══════════════════════════════════════════════════╣
║ Run: 2026-02-19 15:30:00 (Europe/Athens)         ║
║ Mode: standard                                   ║
╚══════════════════════════════════════════════════╝

  SKILLS
  ✅ 20/20 SKILL.md files present
  ✅ 20/20 frontmatter valid
  ✅ 19/19 EVALS.json present
  ✅ All EVALS have 5+ test cases

  DIRECTORIES
  ✅ 16/16 top-level directories present
  ✅ No unexpected top-level directories
  ⚠️  2 subdirectories missing (see details)

  PERMISSIONS
  ✅ /data/auth/ — mode 700
  ❌ /data/clients/ — mode 755 (expected 700)

  DEPENDENCIES
  ✅ jq installed (v1.7.1)
  ✅ curl installed (v8.5.0)
  ✅ openssl installed (v3.0.13)
  ⚠️  OPENCLAW_ENCRYPTION_KEY not set

  BACKUPS
  ✅ Full backup: 2026-02-16 02:00 (3 days ago)
  ✅ Incremental: 2026-02-19 03:00 (12 hours ago)
  ⚠️  Last verification: 2026-02-10 (9 days ago)

  LOCKS
  ✅ No stale process locks

  ENCRYPTION
  ⚠️  Encryption not detected on /data/clients/
  ⚠️  OPENCLAW_ENCRYPTION_KEY not set

  INTEGRITY
  ✅ 847/847 files match stored hashes

═══════════════════════════════════════════════════
  RESULT: 1 FAIL, 4 WARN — review required
  Details: openclaw health-check --all --verbose
═══════════════════════════════════════════════════

JSON Report

Written to /data/reports/system/health-check_{YYYYMMDD}_{HHMMSS}.json:

{
  "timestamp": "2026-02-19T15:30:00Z",
  "mode": "standard",
  "result": "WARN",
  "summary": {
    "pass": 18,
    "fail": 1,
    "warn": 4,
    "skip": 0
  },
  "checks": [
    {
      "category": "skills",
      "check": "skill_files_present",
      "result": "PASS",
      "details": "20/20 SKILL.md files present"
    },
    {
      "category": "permissions",
      "check": "clients_directory_mode",
      "result": "FAIL",
      "expected": "700",
      "actual": "755",
      "remediation": "chmod 700 /data/clients/"
    }
  ]
}

Cron Configuration

{
  "cron": {
    "health-quick": {
      "schedule": "0 */4 * * *",
      "command": "openclaw health-check --quick --save-report",
      "description": "Quick check every 4 hours"
    },
    "health-standard": {
      "schedule": "0 7 * * *",
      "command": "openclaw health-check --standard --save-report",
      "description": "Standard check every morning at 07:00 Athens"
    },
    "health-deep": {
      "schedule": "0 3 * * 0",
      "command": "openclaw health-check --deep --save-report",
      "description": "Deep check every Sunday at 03:00 Athens"
    }
  }
}

Integration with Other Skills

Dashboard Integration

# Health check status feeds into the dashboard system health panel
openclaw dashboard refresh --state system-health
# Dashboard shows: last check time, result, and count of issues

Memory Integration

# Failed health checks are logged as system failures for pattern detection
# Written to /data/memory/failures/ with failure_type: health_check_failed

Meta-Skill Integration

# Health check runs automatically before major operations
openclaw greek monthly-process --pre-flight-check
# Equivalent to: openclaw health-check --quick && openclaw greek monthly-process

Exit Codes

Code Meaning
0 All checks passed (may include WARN)
1 One or more FAIL results
2 Health check itself failed to run (dependency missing, permission error)

常见问题

5 问
答案由本站依据条目字段整理
系统健康检查怎么安装?

在支持 Skills 的客户端里执行:skillhub install xh-ai/system-health-check。安装方式为未标注。装完刷新技能列表或重开客户端。

系统健康检查支持哪些客户端?

资料没有标注宿主客户端,但只要客户端支持 Skills(Claude、Codex、Cursor 等),一般都能加载。

系统健康检查权限高吗,安全吗?

权限级别为未标注权限。安装前先看它会读取或写入哪些内容。

系统健康检查收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于技能本身和所用的模型或客户端,正式使用前按作者说明确认。

系统健康检查用什么许可证?

许可证标注为 未说明(许可证不明确)。商用或二次分发前请到仓库确认条款原文。

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 未标注。