aiskillstore/marketplace / skills/727474430/bug-diagnose-and-root-cause-and-fix
bug-diagnose-and-root-cause-and-fix
Verified【Bug 诊断专用】当遇到程序错误、异常行为、性能问题、功能故障时必须使用本技能。触发关键词:报错、Bug、问题、异常、故障、不工作、诊断、排查、修复。本技能提供系统性根因分析,输出详细诊断报告和可执行修复方案,包含代码位置引用、步骤指导、注意事项。仅提供分析和方案,不直接修改代码。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.6 KB |
| skill-report.json | config | 12.7 KB |
skill-report.json
View on GitHub{
"schema_version": "2.0",
"meta": {
"generated_at": "2026-07-04T17:11:44.909Z",
"slug": "727474430-bug-diagnose-and-root-cause-and-fix",
"source_url": "https://github.com/727474430/cli-rule-swtich/tree/main/templates/claude/jige/skills/bug-diagnose-and-root-cause-and-fix",
"source_ref": "main",
"model": "codex",
"analysis_version": "3.0.0",
"source_type": "community",
"content_hash": "3ca301825e3ba18d373066c0ab17f33bc54a6ecc2c5a7130bf3ada2e614367bd",
"tree_hash": "aec3c8853a799d3728ae10b4bc4ff0ed0379c173eac1aaff7343d16686b9c8ab",
"provenance": {
"model_effective": "codex",
"fallback_chain": [
{
"agent": "codex",
"outcome": "success"
}
],
"analysis_version": "3.0.0"
}
},
"skill": {
"name": "bug-diagnose-and-root-cause-and-fix",
"description": "Guides AI assistants through systematic bug diagnosis, root cause analysis, and repair planning. It asks for evidence, reviews relevant code, and returns a structured fix report without directly changing files.",
"summary": "Diagnose software bugs and produce evidence-based root cause reports with repair steps.",
"icon": "📦",
"version": "1.0.1",
"author": "727474430",
"license": "MIT",
"tags": [
"debugging",
"root-cause-analysis",
"bug-fix-planning",
"code-review",
"diagnostics"
],
"supported_tools": [
"claude",
"codex",
"claude-code"
],
"risk_factors": [
"external_commands"
],
"category": "coding"
},
"security_audit": {
"risk_level": "safe",
"is_blocked": false,
"safe_to_publish": true,
"analysis_status": "ok",
"summary": "The static external command findings are false positives caused by Markdown inline code examples, not executable code. The obfuscation heuristic is also a false positive because the skill file is readable Markdown content with no encoded payload or prompt injection evidence.",
"remediation": [
{
"issue": "Diagnostic command guidance is broad.",
"suggestion": "Clarify that Bash usage should be limited to read-only diagnostic commands and should avoid destructive or secret-printing commands.",
"severity": "low"
}
],
"risk_factor_evidence": [
{
"factor": "external_commands",
"evidence": [
{
"file": "SKILL.md",
"line_start": 60,
"line_end": 60
},
{
"file": "SKILL.md",
"line_start": 69,
"line_end": 69
},
{
"file": "SKILL.md",
"line_start": 76,
"line_end": 76
},
{
"file": "SKILL.md",
"line_start": 97,
"line_end": 97
}
]
}
],
"static_findings": [
{
"id": "external_commands:SKILL.md:60:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 60,
"line_end": 60,
"snippet": "- 引用关键代码片段(使用 `文件路径:行号` 格式)"
},
{
"id": "external_commands:SKILL.md:69:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 69,
"line_end": 69,
"snippet": "- 需要修改的文件:`文件路径`"
},
{
"id": "external_commands:SKILL.md:76:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 76,
"line_end": 76,
"snippet": "- 运行命令:`具体命令`"
},
{
"id": "external_commands:SKILL.md:97:ruby-shell-backtick-execution",
"category": "external_commands",
"severity": "medium",
"pattern": "Ruby/shell backtick execution",
"file": "SKILL.md",
"line_start": 97,
"line_end": 97,
"snippet": "- **引用准确** - 使用 `文件:行号` 格式标注所有相关代码位置"
},
{
"id": "obfuscation:SKILL.md:1:heuristic-high-file-entropy-6-91-bits-possible-b",
"category": "obfuscation",
"severity": "high",
"pattern": "[HEURISTIC] High file entropy (6.91 bits) - possible binary/encrypted content",
"file": "SKILL.md",
"line_start": 1,
"line_end": 1,
"snippet": "File: SKILL.md"
}
],
"finding_verdicts": [
{
"id": "external_commands:SKILL.md:60:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "Line 60 uses Markdown inline code to describe a file path and line number format. It is documentation text, not Ruby or shell execution."
},
{
"id": "external_commands:SKILL.md:69:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "Line 69 uses Markdown inline code around a placeholder file path. There is no executable code, interpreter call, or command substitution."
},
{
"id": "external_commands:SKILL.md:76:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.92,
"reason": "Line 76 shows a report template placeholder for a command to run during verification. The backticks are Markdown formatting, not shell backtick execution."
},
{
"id": "external_commands:SKILL.md:97:ruby-shell-backtick-execution",
"verdict": "false_positive",
"confidence": 0.96,
"reason": "Line 97 uses Markdown inline code to show a citation format. It does not execute commands or interpolate shell output."
},
{
"id": "obfuscation:SKILL.md:1:heuristic-high-file-entropy-6-91-bits-possible-b",
"verdict": "false_positive",
"confidence": 0.88,
"reason": "SKILL.md is plain text Markdown with Chinese prose and normal front matter. The high entropy heuristic appears caused by multilingual text, not encoded or encrypted content."
}
],
"semantic_findings": [],
"critical_findings": [],
"high_findings": [],
"medium_findings": [],
"low_findings": [],
"dangerous_patterns": [],
"files_scanned": 1,
"total_lines": 98,
"audit_model": "codex",
"audited_at": "2026-07-04T17:11:44.909Z",
"agent_auto_install_policy": "allowed",
"manual_install_policy": "allowed"
},
"content": {
"user_title": "Diagnose Bugs and Plan Fixes",
"value_statement": "Software failures often have unclear causes and scattered evidence. This skill guides Claude, Codex, and Claude Code to produce structured diagnosis reports with practical repair steps.",
"seo_keywords": [
"Claude bug diagnosis",
"Codex debugging skill",
"Claude Code root cause analysis",
"AI bug triage",
"software debugging workflow",
"root cause report",
"bug fix planning",
"code diagnostics",
"developer troubleshooting",
"repair guidance"
],
"actual_capabilities": [
"Collects error symptoms, reproduction steps, and missing context before analysis.",
"Guides review of relevant files, functions, logs, and recent code changes.",
"Builds evidence-based hypotheses and validates them against code behavior.",
"Produces root cause explanations with file and line references.",
"Creates step-by-step repair plans with validation commands and expected results.",
"Compares multiple fix options when more than one solution is reasonable."
],
"limitations": [
"Does not directly modify code or apply patches.",
"Requires access to relevant source files, logs, or error messages.",
"May need user follow-up when reproduction steps are incomplete.",
"Command suggestions still need review before running in sensitive environments."
],
"use_cases": [
{
"title": "Triage Production Defects",
"description": "Analyze symptoms, logs, and recent changes to identify the most likely cause of a service defect.",
"target_user": "Backend engineers"
},
{
"title": "Debug Frontend Failures",
"description": "Trace broken UI behavior to component logic, state handling, routing, or integration errors.",
"target_user": "Frontend developers"
},
{
"title": "Prepare Repair Guidance",
"description": "Create clear fix steps and validation checks before assigning implementation work.",
"target_user": "Engineering leads"
}
],
"prompt_templates": [
{
"title": "Basic Bug Diagnosis",
"scenario": "Use when you have an error message and need a structured investigation.",
"prompt": "Use this skill to diagnose this bug. Review the error message, identify likely causes, and produce a root cause report with repair steps."
},
{
"title": "Reproduction-Based Investigation",
"scenario": "Use when a bug has clear steps to reproduce.",
"prompt": "Use this skill to analyze these reproduction steps. Map each step to the relevant code path, explain where behavior diverges, and propose a fix plan."
},
{
"title": "Regression Root Cause",
"scenario": "Use when a bug appeared after a recent change.",
"prompt": "Use this skill to investigate this regression. Check recent changes, identify the root cause, and compare at least two repair options with tradeoffs."
},
{
"title": "Complex Cross-System Failure",
"scenario": "Use when the failure involves multiple services, packages, or runtime conditions.",
"prompt": "Use this skill for a deep diagnosis. Build hypotheses, identify evidence needed for each one, cite relevant files and lines, and produce a prioritized remediation plan."
}
],
"output_examples": [
{
"input": "A payment callback returns success, but the order remains pending after deployment.",
"output": "A diagnosis report that identifies the likely state transition failure, cites the handler and order update path, and lists repair and validation steps."
},
{
"input": "A React form submits twice when users press Enter after selecting an option.",
"output": "A root cause explanation that traces event handling, identifies duplicate submit triggers, and recommends a focused fix with browser validation checks."
},
{
"input": "An API endpoint became slow after a package upgrade.",
"output": "A performance diagnosis that compares recent changes, identifies the slow dependency path, and proposes measurement and rollback options."
}
],
"best_practices": [
"Provide exact error messages, stack traces, and reproduction steps before asking for diagnosis.",
"Share relevant files or logs so conclusions can cite concrete evidence.",
"Review suggested commands and fixes before using them in production systems."
],
"anti_patterns": [
"Do not ask for a fix without providing the observed behavior and expected behavior.",
"Do not treat the first hypothesis as final without checking supporting evidence.",
"Do not run suggested verification commands blindly in sensitive environments."
],
"faq": [
{
"question": "Does this skill change my code?",
"answer": "No. It produces diagnosis reports and repair guidance, but it does not directly edit files."
},
{
"question": "Which tools can use this skill?",
"answer": "It is structured for Claude, Codex, and Claude Code workflows."
},
{
"question": "What information should I provide?",
"answer": "Provide symptoms, error messages, reproduction steps, relevant files, and recent changes."
},
{
"question": "Can it handle performance issues?",
"answer": "Yes. It can guide performance diagnosis when timing data, traces, or relevant code paths are available."
},
{
"question": "Will it browse documentation automatically?",
"answer": "No. The skill says to identify needed documentation queries in the report when outside context is required."
},
{
"question": "How should I use the output?",
"answer": "Use the report to review evidence, choose a fix option, implement changes, and run the listed validation steps."
}
]
},
"file_structure": [
{
"name": "SKILL.md",
"type": "file",
"path": "SKILL.md",
"lines": 98
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |