Token导航 LogoToken导航TokenDH.com
待分类只读github未标认证来源可访问clear审计提醒

validate-system验证系统

Agent Skill

validate-system 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。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

261

周安装

11

GitHub Stars

2

下载量

92
CodexClaudeCursorGemini CLI

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

3

许可证

MIT

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:validate-system(验证系统)
来源仓库:https://github.com/abdullahbeam/nexus-design-abdullah
仓库路径:skills/validate-system
安装命令:
npx skills add https://github.com/abdullahbeam/nexus-design-abdullah --skill validate-system
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/abdullahbeam/nexus-design-abdullah --skill validate-system

简介

validate-system 全面检查系统配置、服务状态和资源使用情况。

  • 适用于运维监控、故障排查或系统健康度评估场景。
  • 可识别内存泄漏、端口冲突、服务宕机等常见问题征兆。
  • 部分深层问题可能需要重启或人工干预才能彻底解决。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Validate System

Check system integrity, auto-fix issues, validate maps, and run optional Python validation hooks.

Purpose

The validate-system skill performs comprehensive system health checks, validates structural integrity, checks navigation map accuracy, runs optional Python validation hooks, and attempts to auto-fix common issues. Use this skill when navigation seems stale, after manual file changes, or as periodic maintenance.

Key Features:

  • Comprehensive Checks: Core files, structure, memory, navigation, projects, skills
  • Map Integrity Validation: Ensures skill-map.md and project-map.md are accurate
  • Python Hooks: Runs automated validators (optional, graceful skip if unavailable)
  • Auto-Fix: Regenerates navigation, recreates templates, repairs structure
  • Detailed Reporting: Clear report of checks, issues, and fixes

Time Estimate: <10 seconds for full validation


Workflow

Step 1: Initialize TodoList

Create TodoWrite with all validation steps:

- [ ] Check core files
- [ ] Check folder structure
- [ ] Check memory files
- [ ] Check navigation files
- [ ] Validate projects
- [ ] Validate skills
- [ ] Check map integrity
- [ ] Run Python hooks (if available)
- [ ] Auto-fix issues
- [ ] Generate report
- [ ] Display report
- [ ] Close session to save progress

Mark tasks complete as you finish each step.

Step 2: Execute Validation Workflow

This skill performs comprehensive system validation with the following checks:

  1. Core Files - Verify critical system files exist
  2. Folder Structure - Check required directories
  3. Memory Files - Validate Memory/ content
  4. Navigation Files - Check framework-map.md and skill-map.md
  5. Projects - Validate all projects in 02-Projects/
  6. Skills - Validate all skills in 00-system/Skills/
  7. Map Integrity - Ensure maps match actual files
  8. Python Hooks - Run validation hooks (if available)
  9. Auto-Fix - Repair common issues automatically
  10. Report - Generate and display comprehensive report

Detailed validation steps: See references/validation-checks.md

Report templates: See references/report-templates.md

Time Estimate: <10 seconds for full validation

Final Step: Close Session

Automatically trigger the close-session skill:

Auto-triggering close-session to save progress...

This ensures all validation results and any auto-fixes are properly saved to memory.


Error Handling

Critical Files Missing

Scenario: framework-map.md, orchestrator.md, or claude.md missing

Action:

  • Report as CRITICAL ERROR
  • Cannot auto-fix (system corrupted)
  • Suggest: "Your system appears corrupted. Please restore from backup or reinstall Nexus-v3."
  • Do NOT proceed with auto-fixes (unsafe)

Python Hook Crashes

Scenario: Python hook script crashes or returns invalid JSON

Action:

  • Catch error gracefully
  • Report: "Hook {name} crashed: {error}"
  • Add to issues list
  • Continue with other hooks (don't block validation)

Cannot Write to Files

Scenario: Auto-fix fails due to file permission issues

Action:

  • Report: "Cannot write to {file}: {error}"
  • Add to manual fix list
  • Provide manual instructions
  • Continue with other auto-fixes

Corrupted Memory Files

Scenario: Memory/ files exist but are not valid markdown

Action:

  • Report: "{file} is corrupted (invalid markdown)"
  • Offer: "Replace with empty template? This will lose current content."
  • Wait for user confirmation
  • IF confirmed → Replace with template
  • IF declined → Add to manual fix list

Empty Projects/ Folder

Scenario: Projects/ exists but is empty (only during first-time setup)

Action:

  • Report: "Projects/ folder is empty (no projects yet)"
  • Note: "This is normal for new users."
  • Suggest: "Say 'create project' to start your first project!"
  • Mark as INFO (not an error)

Empty Skills/ Folder

Scenario: Skills/ exists but is empty

Action:

  • Report: "No user skills found (Skills/ folder empty)"
  • Note: "This is normal if you haven't created any skills yet."
  • Suggest: "Run create-skill to add your first skill!"
  • Mark as INFO (not an error)

Notes

When to Run validate-system:

  • After manual file edits
  • When navigation seems stale or inaccurate
  • After accidentally deleting files
  • As periodic maintenance (weekly/monthly)
  • When system behavior seems off

Auto-Fix Safety:

  • Only fixes non-destructive issues
  • Never deletes user content
  • Creates templates, regenerates navigation
  • Critical issues require manual intervention

Python Hooks:

  • Completely optional (system works without them)
  • Provide deeper automated validation
  • Gracefully skip if Python not available
  • Custom hooks can be added to 00-system/hooks/

Hook JSON Format:

{
  "valid": true/false,
  "errors": [
    {"file": "path", "line": 42, "message": "error description", "severity": "error"}
  ],
  "warnings": [
    {"file": "path", "line": 15, "message": "warning description", "severity": "warning"}
  ]
}

Map Integrity Importance:

  • skill-map.md MUST match Skills/ folder
  • 02-projects/project-map.md MUST match Projects/ folder
  • Dead links break navigation
  • close-session maintains integrity automatically

Validation Frequency:

  • Run when suspicious: anytime something feels off
  • Run after manual changes: edited files outside skills
  • Run periodically: weekly or monthly maintenance
  • NOT needed after normal skill/project usage (close-session handles it)

Integration with close-session:

  • close-session automatically maintains map integrity
  • validate-system catches issues between sessions
  • Both work together for system health

Remember: validate-system is your system health check. Run it when things seem off, and let it auto-fix what it can!

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

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

Claude Code

31.07%
按下载量换算29

Antigravity

24.25%
按下载量换算22

windsurf

16.89%
按下载量换算16

Codex

13%
按下载量换算12

OpenCode

8.53%
按下载量换算8

Gemini CLI

3.91%
按下载量换算4

安全审计

Gen Agent Trust Hub

可疑

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills