debug — for Claude Code epost_agent_kit, community, for Claude Code, ide skills, skill-discovery, Unified, Command, platform-specific, issues, automatic

v1.0.0

关于此技能

适合需要高级平台特定问题检测和错误分析功能的AI代理。 本地化技能摘要: (ePost) Debug issues — auto-detects platform # Debug — Unified Debug Command Debug platform-specific issues with automatic platform detection. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

功能特性

Debug — Unified Debug Command
Debug platform-specific issues with automatic platform detection.
Platform Detection
Detect platform per skill-discovery protocol.
Route to platform-specific agent (read-only investigation tools preferred)

# 核心主题

Klara-copilot Klara-copilot
[2]
[0]
更新于: 3/9/2026

技能概览

先看适用场景、限制条件和安装路径,再决定是否继续深入。

适合需要高级平台特定问题检测和错误分析功能的AI代理。 本地化技能摘要: (ePost) Debug issues — auto-detects platform # Debug — Unified Debug Command Debug platform-specific issues with automatic platform detection. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

核心价值

赋予代理自动检测平台特定问题、分析错误上下文并使用只读调查工具和技能发现协议建议修复的能力,提供统一的调试命令以实现高效调试。

适用 Agent 类型

适合需要高级平台特定问题检测和错误分析功能的AI代理。

赋予的主要能力 · debug

使用自动平台检测进行平台特定问题的调试
分析错误上下文并找出根本原因
为平台特定问题建议修复,而不自动应用更改

! 使用限制与门槛

  • 需要平台检测功能
  • 仅限于只读调查工具
  • 不自动应用修复,只建议修复

关于来源内容

The section below comes from the upstream repository. Use it as supporting material alongside the fit, use-case, and installation summary on this page.

实验室 Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

常见问题与安装步骤

以下问题与步骤与页面结构化数据保持一致,便于搜索引擎理解页面内容。

? FAQ

debug 是什么?

适合需要高级平台特定问题检测和错误分析功能的AI代理。 本地化技能摘要: (ePost) Debug issues — auto-detects platform # Debug — Unified Debug Command Debug platform-specific issues with automatic platform detection. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

如何安装 debug?

运行命令:npx killer-skills add Klara-copilot/epost_agent_kit/debug。支持 Cursor、Windsurf、VS Code、Claude Code 等 19+ IDE/Agent。

debug 适用于哪些场景?

典型场景包括:使用自动平台检测进行平台特定问题的调试、分析错误上下文并找出根本原因、为平台特定问题建议修复,而不自动应用更改。

debug 支持哪些 IDE 或 Agent?

该技能兼容 Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer。可使用 Killer-Skills CLI 一条命令通用安装。

debug 有哪些限制?

需要平台检测功能;仅限于只读调查工具;不自动应用修复,只建议修复。

安装步骤

  1. 1. 打开终端

    在你的项目目录中打开终端或命令行。

  2. 2. 执行安装命令

    运行:npx killer-skills add Klara-copilot/epost_agent_kit/debug。CLI 会自动识别 IDE 或 AI Agent 并完成配置。

  3. 3. 开始使用技能

    debug 已启用,可立即在当前项目中调用。

! 来源说明

此页面仍可作为安装与查阅参考。继续使用前,请结合上方适用场景、限制条件和上游仓库说明一起判断。

Upstream Repository Material

The section below comes from the upstream repository. Use it as supporting material alongside the fit, use-case, and installation summary on this page.

Upstream Source

debug

安装 debug,这是一款面向AI agent workflows and automation的 AI Agent Skill。查看功能、使用场景、限制条件与安装命令。

SKILL.md
Readonly
Upstream Repository Material
The section below comes from the upstream repository. Use it as supporting material alongside the fit, use-case, and installation summary on this page.
Upstream Source

Debug — Unified Debug Command

Debug platform-specific issues with automatic platform detection.

Platform Detection

Detect platform per skill-discovery protocol.

Execution

  1. Detect platform
  2. Route to platform-specific agent (read-only investigation tools preferred)
  3. Analyze error context, gather logs, identify root cause
  4. Explain root cause and suggest fix (do NOT auto-apply fix — that's /fix)

Expertise

Systematic Debugging

  1. Understand: What's the symptom?
  2. Reproduce: Can you reproduce it?
  3. Isolate: What's the minimal case?
  4. Analyze: What's actually happening?
  5. Hypothesize: What could cause this?
  6. Verify: Does the fix work?

Log Analysis

  • Parse error messages
  • Follow stack traces
  • Identify log patterns
  • Contextual logging

Stack Trace Interpretation

  • Read top-down
  • Identify root cause frame
  • Distinguish cause from symptom
  • Async stack traces

Reproduction Strategies

  • Minimal reproduction
  • Environment matching
  • Data setup
  • Step reproduction

Root Cause Analysis

See problem-solving for root cause analysis techniques (5 Whys, bisection, inversion).

Fix Validation

  • Regression testing
  • Edge case testing
  • Performance impact
  • Side effect analysis

Patterns

Debug Logging

typescript
1console.log('[Feature]', { variable, state }); 2console.debug('[Debug]', value); 3console.error('[Error]', error);

Error Boundaries

typescript
1class ErrorBoundary extends Component { 2 componentDidCatch(error, errorInfo) { 3 console.error('Error caught:', error, errorInfo); 4 } 5}

Debug Mode

typescript
1const DEBUG = process.env.DEBUG === 'true'; 2if (DEBUG) console.debug('Debug info');

Structured Logging

typescript
1logger.info('User action', { 2 action: 'login', 3 userId: user.id, 4 timestamp: Date.now() 5});

Common Issues

TypeScript

  • Type mismatches
  • Missing type imports
  • Any type issues
  • Generic constraints

React

  • Stale closures
  • Missing dependencies
  • Re-render loops
  • State timing

Async

  • Race conditions
  • Promise rejection
  • Missing await
  • Callback hell

Defense-in-Depth Patterns

Validation Layers

  1. Input validation (reject invalid early)
  2. Business logic validation (enforce invariants)
  3. Output validation (verify results before return)

Error Handling Strategy

  • Catch: Only where you can handle meaningfully
  • Transform: Convert to domain-specific errors
  • Log: Include context (not just message)
  • Propagate: Let upstream handle if you can't

Assertion vs Exception

  • Assertions: Programmer errors (should never happen)
  • Exceptions: Runtime problems (can happen legitimately)

State Diagram Tracing

When debugging state-related bugs (unexpected transitions, stuck states, race conditions):

  1. Draw the ACTUAL state machine from code — read every if/switch/state= and extract what ACTUALLY happens
  2. Draw the EXPECTED state machine from requirements or docs
  3. Overlay and diff — mismatches reveal the bug:
    • Missing transitions (no path from state A to B)
    • Unguarded transitions (state changes without preconditions)
    • Dead states (reachable but no exit — component gets "stuck")
    • Race conditions (two transitions competing for same state)
ACTUAL:   [LOADING] ──(timeout)──▸ [LOADING]     ← stuck! no error path
EXPECTED: [LOADING] ──(timeout)──▸ [ERROR] ──(retry)──▸ [LOADING]
MISSING:  timeout → ERROR transition

Applies to: React useState/useReducer, iOS view lifecycle, Android Compose state, async/Promise chains, WebSocket connections.

See plan/references/state-machine-guide.md for notation and common patterns.

Verification Checklist

  • Symptom reproduced consistently
  • Root cause identified and documented
  • Fix applied and tested
  • No new issues introduced
  • Edge cases considered
  • Similar issues checked elsewhere in codebase

Tools

  • Browser DevTools (breakpoints, profiling)
  • Node debugger (--inspect)
  • Console logging (structured)
  • Source maps (correct line numbers)
  • Test suite (regression testing)

Debugging Discipline

IRON LAW: NO FIXES WITHOUT ROOT CAUSE FIRST.

Applying a fix before identifying root cause is not debugging — it is guessing. Guesses compound into technical debt.

See verification-before-completion skill for anti-rationalization table, red flags, and the full verification gate protocol.

Sub-Skill Routing

IntentSub-SkillWhen
Fix broken codefix/fix, "fix this", error/crash/failing
Fix deeplyfix-deep/fix-deep, complex multi-file bugs
Fix CI pipelinefix-ci/fix-ci, CI/CD failures, build pipeline
Fix UI issuesfix-ui/fix-ui, visual bugs, layout broken
  • knowledge-retrieval — Knowledge storage format and docs/ directory
  • knowledge-capture — Post-task capture workflow
  • problem-solving — Root cause analysis techniques
  • error-recovery — Error handling and recovery patterns
  • verification-before-completion — Verify fixes before claiming done
  • auto-improvement — Error metrics auto-captured by session-metrics hook on Stop

References

  • references/debugging-flow.dot — Authoritative debugging process flowchart
  • references/condition-based-waiting.md — Patterns for replacing sleep() with condition polling

<issue>$ARGUMENTS</issue>

IMPORTANT: Analyze the skills catalog and activate needed skills for the detected platform.

相关技能

寻找 debug 的替代方案 (Alternative) 或可搭配使用的同类 community Skill?探索以下相关开源技能。

查看全部

openclaw-release-maintainer

Logo of openclaw
openclaw

本地化技能摘要: 🦞 # OpenClaw Release Maintainer Use this skill for release and publish-time workflow. It covers ai, assistant, crustacean workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

333.8k
0
AI

widget-generator

Logo of f
f

本地化技能摘要: Generate customizable widget plugins for the prompts.chat feed system # Widget Generator Skill This skill guides creation of widget plugins for prompts.chat . It covers ai, artificial-intelligence, awesome-list workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf

149.6k
0
AI

flags

Logo of vercel
vercel

本地化技能摘要: The React Framework # Feature Flags Use this skill when adding or changing framework feature flags in Next.js internals. It covers blog, browser, compiler workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

138.4k
0
浏览器

pr-review

Logo of pytorch
pytorch

本地化技能摘要: Usage Modes No Argument If the user invokes /pr-review with no arguments, do not perform a review . It covers autograd, deep-learning, gpu workflows. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

98.6k
0
开发者工具