codex-audit

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Codex Audit Skill

Codex审计技能

LIBRARY-FIRST PROTOCOL (MANDATORY)

库优先协议（强制性）

Before writing ANY code, you MUST check:

在编写任何代码之前，你必须检查：

Step 1: Library Catalog

步骤1：库目录

Location:
```
.claude/library/catalog.json
```
If match >70%: REUSE or ADAPT

位置：
```
.claude/library/catalog.json
```
如果匹配度>70%：复用或适配

Step 2: Patterns Guide

步骤2：模式指南

Location:

.claude/docs/inventories/LIBRARY-PATTERNS-GUIDE.md

If pattern exists: FOLLOW documented approach

位置：

.claude/docs/inventories/LIBRARY-PATTERNS-GUIDE.md

如果模式已存在：遵循文档记录的方法

Step 3: Existing Projects

步骤3：现有项目

Location:
```
D:\Projects\*
```
If found: EXTRACT and adapt

位置：
```
D:\Projects\*
```
如果找到：提取并适配

Decision Matrix

决策矩阵

Match	Action
Library >90%	REUSE directly
Library 70-90%	ADAPT minimally
Pattern exists	FOLLOW pattern
In project	EXTRACT
No match	BUILD (add to library after)

匹配度	操作
库匹配度>90%	直接复用
库匹配度70-90%	最小程度适配
模式已存在	遵循模式
存在于项目中	提取
无匹配项	构建（之后添加到库中）

Purpose

用途

Route auditing and debugging tasks to Codex CLI when:

Autonomous iteration is needed (test-fix-retest loops)
Sandboxed execution required for safety
Rapid prototyping without approval overhead

在以下场景将审计和调试任务路由至Codex CLI：

需要自主迭代（测试-修复-重测循环）
出于安全考虑需要沙箱执行
无需审批流程的快速原型开发

Unique Capability

独特能力

What Codex Does Better:

Fully autonomous execution (no approval needed mid-task)
Sandboxed isolation (no network, CWD only)
Iterative debugging loops
GPT-5-Codex optimized for agentic coding

Codex的优势：

完全自主执行（任务中途无需审批）
沙箱隔离（无网络访问，仅当前工作目录）
迭代式调试循环
GPT-5-Codex针对智能体编码优化

When to Use

使用场景

Perfect For:

适用场景：

Automated test fixing
Code auditing in isolation
Rapid prototyping of features
Refactoring with test verification
Build failure recovery
Security scanning in sandbox

自动化测试修复
隔离环境下的代码审计
功能快速原型开发
带测试验证的重构
构建失败恢复
沙箱内安全扫描

Don't Use When:

不适用场景：

Need network access (sandbox disables it)
Need to access files outside CWD
Production debugging (use Claude with oversight)
Complex multi-file coordination

需要网络访问（沙箱已禁用）
需要访问当前工作目录（CWD）以外的文件
生产环境调试（使用带监督的Claude）
复杂多文件协调

Usage

使用方法

Basic Audit

基础审计

bash

/codex-audit "Find and fix all type errors" --context src/

bash

/codex-audit "Find and fix all type errors" --context src/

Test Fixing

测试修复

bash

/codex-audit "Fix failing tests" --context tests/ --max-iterations 10

bash

/codex-audit "Fix failing tests" --context tests/ --max-iterations 10

Prototyping

原型开发

bash

/codex-audit "Build REST API with CRUD endpoints" --context .

bash

/codex-audit "Build REST API with CRUD endpoints" --context .

Command Pattern

命令格式

bash

bash scripts/multi-model/codex-audit.sh "<task>" "<context>" "<task_id>" "<max_iterations>"

bash

bash scripts/multi-model/codex-audit.sh "<task>" "<context>" "<task_id>" "<max_iterations>"

Safety Constraints

安全限制

Constraint	Value
Network	DISABLED
File Access	CWD only
Isolation	macOS Seatbelt / Docker
Max Iterations	5 (configurable)

限制项	值
网络	禁用
文件访问	仅当前工作目录（CWD）
隔离方式	macOS Seatbelt / Docker
最大迭代次数	5（可配置）

Memory Integration

内存集成

Results stored to Memory-MCP:

Key:
```
multi-model/codex/audit/{task_id}
```
Tags: WHO=codex-cli, WHY=audit

结果存储至Memory-MCP：

键：
```
multi-model/codex/audit/{task_id}
```
标签：WHO=codex-cli, WHY=audit

Output Format

输出格式

json

{
  "raw_output": "Audit findings...",
  "metrics": {
    "files_analyzed": 15,
    "findings_count": 7,
    "fixes_applied": 5
  },
  "context_path": "src/",
  "sandbox_mode": true
}

json

{
  "raw_output": "Audit findings...",
  "metrics": {
    "files_analyzed": 15,
    "findings_count": 7,
    "fixes_applied": 5
  },
  "context_path": "src/",
  "sandbox_mode": true
}

Handoff to Claude

交接至Claude

After Codex audit completes:

Findings stored in Memory-MCP
Claude agents review findings
Apply or escalate based on severity

javascript

// Claude agent reads Codex audit
const audit = memory_retrieve("multi-model/codex/audit/{task_id}");
if (audit.metrics.findings_count > 0) {
  Task("Reviewer", `Review findings: ${audit.raw_output}`, "reviewer");
}

Codex审计完成后：

审计结果存储至Memory-MCP
Claude智能体审核结果
根据严重程度应用修复或升级处理

javascript

// Claude agent reads Codex audit
const audit = memory_retrieve("multi-model/codex/audit/{task_id}");
if (audit.metrics.findings_count > 0) {
  Task("Reviewer", `Review findings: ${audit.raw_output}`, "reviewer");
}

Integration with Audit Pipeline

与审计流水线集成

bash

undefined

bash

undefined

Phase 1: Theater detection (Claude)

/theater-detection-audit

Phase 2: Functionality audit (Codex)

/codex-audit "Verify all functions work" --context src/

Phase 3: Style audit (Claude)

/style-audit

undefined

/style-audit

undefined