troubleshooting

Diagnose root causes from logs and code context in pipelines and deployments.

159|47|Updated Feb 26, 2026
One-click install
npx skills add https://github.com/davidYichengWei/agentic-engineering-framework --skill troubleshooting-davidyichengwei
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: troubleshooting
Source: https://github.com/davidYichengWei/agentic-engineering-framework/tree/main/skills/troubleshooting
Command: npx skills add https://github.com/davidYichengWei/agentic-engineering-framework --skill troubleshooting-davidyichengwei

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

该 Skill 指导用户在面对编译错误、运行时异常、单元测试失败、CI/CD 流水线报错或现网告警等问题时,基于证据进行系统性的排查与根因定位,避免凭直觉猜测。

Core Features & Use Cases

  • 铁律原则:禁止假设,所有结论必须基于代码与日志证据。
  • 排查记录模板:提供 reference/troubleshooting-log-template.md 规范排查过程的记录。
  • 信息驱动排查:要求提供关键日志、环境信息和代码上下文,并可通过 codebase-researcher 调研相关代码上下文。
  • 操作流程:按假设-验证循环推进,实时更新待确认点,3+轮结束时扩大信息范围。

Quick Start

描述你遇到的故障现象并提供相关日志,我将引导你完成基于证据的根因分析。

Frequently Asked Questions about troubleshooting

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find the root cause of a CI/CD pipeline failure or production incident?

To find the root cause of a CI/CD pipeline failure or production incident, provide relevant logs and code context for an evidence-based diagnosis process that eliminates guesswork and pinpoints exact triggers.

What is the best way to debug runtime exceptions and compile errors using codebase context?

The best way to debug runtime exceptions and compile errors is applying a no-assumptions rule that requires user-provided logs and codebase context to validate hypotheses and output explicit root-cause statements.

How to systematically troubleshoot unit test failures without guessing?

Systematically troubleshoot unit test failures by following a hypothesis-verification loop that analyzes logs and code context, expanding the information scope after three or more unresolved iterations.

Do I need to provide specific environment information for log analysis and root-cause diagnosis?

Yes, you need to provide specific environment information and logs for log analysis and root-cause diagnosis, because the process enforces a strict no-assumptions rule requiring concrete evidence for all conclusions.

When should I expand the information scope during issue diagnosis?

You should expand the information scope during issue diagnosis when a hypothesis-verification loop ends after three or more rounds without confirming a root cause, requiring broader codebase context research.