with one click
debug-analysis
深度调试分析 - 使用 Codex 本地证据检索、命令验证和只读 child agent 进行系统性根因诊断。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Menu
深度调试分析 - 使用 Codex 本地证据检索、命令验证和只读 child agent 进行系统性根因诊断。
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
AI-native code annotation protocol that encodes intent, risk, dependencies, constraints, and test expectations in machine-parseable comments.
Professional Codex-native multi-agent code review with confidence scoring
架构规划专家 - 使用 Codex 本地上下文、仓库证据和只读 child agent 进行大型系统架构设计与规划。
Backend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js API routes.
Imported from everything-codex command checkpoint
Universal coding standards, best practices, and patterns for TypeScript, JavaScript, React, and Node.js development.
| name | debug-analysis |
| description | 深度调试分析 - 使用 Codex 本地证据检索、命令验证和只读 child agent 进行系统性根因诊断。 |
| auto_invoke | true |
| tags | ["debug","analysis","troubleshooting","codex-native"] |
| version | 0.2.0 |
| source | fork |
| checksum | 9f7edc9f189455069a8701e56b2c24f9b642d4308f9536aed2430fd95d37635a |
| updated_at | "2026-05-29T03:50:00.000Z" |
| intent | debugging |
| layer | research |
使用当前 Codex 会话、仓库检索、可复现命令和只读 child agent 完成根因分析。不调用外部模型 CLI、不依赖长上下文外部包装器。
手动触发:
使用 debug-analysis 分析:<问题描述>
收集并确认:
优先使用本地证据:
rg -n "<error-keyword>|<symbol>|<route>" .
rg --files | rg "<suspected-area>"
git diff --stat
git log --oneline -5
输出 3-5 个根因假设,每个假设包含:
当问题跨模块、影响面不清或需要独立视角时,派发只读 child agent:
spawn_agent(agent_type="explorer", message="<trace prompt>")
wait_agent
close_agent
推荐 prompt 框架:
Your task is to trace a bug from evidence. Read-only.
<agent-instructions>
Inputs:
- Symptom: <错误现象>
- Logs: <关键日志>
- Suspected files: <文件列表>
- Reproduction command: <命令>
Trace the real execution path. Cite exact files, symbols, and lines where possible.
Return:
1. Most likely root cause
2. Evidence chain
3. Minimal verification command
4. Risky assumptions
5. Files likely needing changes
</agent-instructions>
如果需要严苛复核,可再派发:
spawn_agent(agent_type="reviewer", message="<review the diagnosis>")
按问题类型选择最小验证:
## Debug Analysis
### Symptom
[用户可见问题]
### Evidence
- [file:line or command output summary]
### Root Cause
[最可能根因,带证据链]
### Verification
- Command: `<command>`
- Result: [pass/fail/key output]
### Fix Direction
1. [最小修复路径]
2. [需要补测的地方]
### Remaining Risk
[未验证假设或外部依赖]
如果 child agent 不可用,在当前回复中用 [EXPLORER]、[REVIEWER]、[SYNTHESIS] 三段自演同样流程,但仍必须基于本地文件、日志和命令验证。