Ralph Skill Review Loop

Ralph技能审查循环

Overview

概述

A meta-skill that uses the Ralph Wiggum technique to review and improve the Ralph Wiggum prompt generator skills themselves. Runs a continuous improvement loop until skills pass review twice consecutively with no recommendations.

一种元技能，使用Ralph Wiggum技术来审查和改进Ralph Wiggum提示生成器技能本身。运行持续改进循环，直到技能连续两次通过审查且无任何改进建议。

Quick Start

快速开始

Copy and run this prompt in a Ralph loop:

bash

/ralph-wiggum:ralph-loop "[paste prompt below]" --completion-promise "RALPH_SKILLS_PERFECTED" --max-iterations 100

复制并在Ralph循环中运行此提示：

bash

/ralph-wiggum:ralph-loop "[paste prompt below]" --completion-promise "RALPH_SKILLS_PERFECTED" --max-iterations 100

THE SELF-IMPROVING REVIEW LOOP PROMPT

自我改进审查循环提示

markdown

undefined

markdown

undefined

Task: Self-Improving Review of Ralph Wiggum Skills

任务：Ralph Wiggum技能的自我改进审查

Objective

目标

Review and improve the Ralph Wiggum prompt generator skills until they pass two consecutive reviews with zero improvement recommendations.

审查并改进Ralph Wiggum提示生成器技能，直到它们连续两次通过审查且无任何改进建议。

Target Skills

目标技能

ralph-prompt-builder (Master orchestrator)
ralph-prompt-single-task (Single task generator)
ralph-prompt-multi-task (Multi-task generator)
ralph-prompt-project (Project generator)
ralph-prompt-research (Research generator)

Location: .claude/skills/ralph-prompt-*/SKILL.md

ralph-prompt-builder（主编排器）
ralph-prompt-single-task（单任务生成器）
ralph-prompt-multi-task（多任务生成器）
ralph-prompt-project（项目生成器）
ralph-prompt-research（研究生成器）

位置：.claude/skills/ralph-prompt-*/SKILL.md

Reference Materials

参考资料

RALPH-WIGGUM-TECHNIQUE-COMPREHENSIVE-RESEARCH.md (12,000+ words of best practices)
skill-builder-package/research/ (skill building best practices)
skill-builder-package/examples/ (production skill patterns)

RALPH-WIGGUM-TECHNIQUE-COMPREHENSIVE-RESEARCH.md（12000+字的最佳实践内容）
skill-builder-package/research/（技能构建最佳实践）
skill-builder-package/examples/（生产级技能模式）

STATE MANAGEMENT

状态管理

Required State Files

所需状态文件

Create these files to track progress:

RALPH_REVIEW_STATE.json:

json

{
  "current_iteration": 1,
  "consecutive_clean_reviews": 0,
  "skills_reviewed": [],
  "improvements_made": [],
  "last_review_timestamp": "",
  "status": "IN_PROGRESS"
}

RALPH_REVIEW_LOG.md:

markdown

undefined

创建以下文件来跟踪进度：

RALPH_REVIEW_STATE.json:

json

{
  "current_iteration": 1,
  "consecutive_clean_reviews": 0,
  "skills_reviewed": [],
  "improvements_made": [],
  "last_review_timestamp": "",
  "status": "IN_PROGRESS"
}

RALPH_REVIEW_LOG.md:

markdown

undefined

Ralph Skills Review Log

Ralph技能审查日志

Iteration History

迭代历史

[Append each iteration's findings here]

---

[在此处追加每次迭代的发现]

---

STEP 1: ORIENTATION (Every Iteration)

步骤1：定位（每次迭代）

Read current state:

bash

cat RALPH_REVIEW_STATE.json
cat RALPH_REVIEW_LOG.md | tail -50
git log --oneline -5
ls -la .claude/skills/ralph-prompt-*/

Check: How many consecutive clean reviews do we have?

If 2 or more: Output <promise>RALPH_SKILLS_PERFECTED</promise>
If less than 2: Continue to Step 2

读取当前状态：

bash

cat RALPH_REVIEW_STATE.json
cat RALPH_REVIEW_LOG.md | tail -50
git log --oneline -5
ls -la .claude/skills/ralph-prompt-*/

检查：我们已有多少次连续无问题审查？

如果达到2次或更多：输出 <promise>RALPH_SKILLS_PERFECTED</promise>
如果少于2次：继续步骤2

STEP 2: COMPREHENSIVE SKILL REVIEW

步骤2：全面技能审查

Review Framework

审查框架

For EACH skill in ralph-prompt-*, evaluate against:

针对每个ralph-prompt-*技能，从以下维度评估：

2.1 Ralph Technique Alignment (from research)

2.1 Ralph技术对齐度（来自研究内容）

Clear completion criteria defined
Includes self-verification commands
Has TDD/iteration approach
Includes "If Stuck" guidance
Uses <promise> completion tags correctly
Recommends appropriate max-iterations
Follows "deterministically bad" philosophy (failures are fixable)

定义了清晰的完成标准
包含自我验证命令
采用TDD/迭代方法
包含“遇到问题时”的指导
正确使用<promise>完成标签
推荐了合适的max-iterations值
遵循“确定性可修复”理念（问题均可被解决）

2.2 Skill Structure Quality

2.2 技能结构质量

YAML frontmatter complete (name, description with triggers)
Progressive disclosure (overview → details → examples)
Quick Start section exists and is actionable
Examples are realistic and complete
Best practices section included
Integration with Ralph loop documented

YAML前置元数据完整（名称、带触发条件的描述）
采用渐进式披露（概述 → 细节 → 示例）
存在Quick Start部分且内容可操作
示例真实且完整
包含最佳实践部分
记录了与Ralph循环的集成方式

2.3 Content Completeness

2.3 内容完整性

All sections properly filled (no placeholders)
Examples match the skill type
Verification commands are real and runnable
Edge cases addressed
Cross-references to related skills

所有部分均已正确填充（无占位符）
示例与技能类型匹配
验证命令真实可运行
考虑到了边缘情况
包含与相关技能的交叉引用

2.4 Prompt Template Quality

2.4 提示模板质量

Templates follow research best practices
Success criteria are measurable
Phase structure is clear (for multi-phase)
State tracking included
Progress tracking pattern included

模板遵循研究中的最佳实践
成功标准可衡量
阶段结构清晰（针对多阶段任务）
包含状态跟踪
包含进度跟踪模式

Review Process

审查流程

For each skill:

Read the SKILL.md file completely
Compare against RALPH-WIGGUM-TECHNIQUE-COMPREHENSIVE-RESEARCH.md
Check against all 16 criteria above
Document findings in REVIEW_FINDINGS.md

针对每个技能：

完整读取SKILL.md文件
与RALPH-WIGGUM-TECHNIQUE-COMPREHENSIVE-RESEARCH.md进行对比
对照上述16项标准检查
将发现记录到REVIEW_FINDINGS.md中

Review Output Format

审查输出格式

Create/update REVIEW_FINDINGS.md:

markdown

undefined

创建/更新REVIEW_FINDINGS.md：

markdown

undefined

Review Findings - Iteration [N]

审查发现 - 迭代[N]

Summary

摘要

Skills reviewed: [count]
Total issues found: [count]
Critical issues: [count]
Improvements needed: [count]

已审查技能：[数量]
发现的总问题数：[数量]
严重问题：[数量]
需要改进的项：[数量]

ralph-prompt-builder

Passing

通过项

Criterion that passes

符合的标准

Issues Found

发现的问题

[CRITICAL/HIGH/MEDIUM/LOW] Issue description
- Location: [section/line]
- Current: [what exists]
- Should be: [what it should be]
- Fix: [specific fix]

[CRITICAL/HIGH/MEDIUM/LOW] 问题描述
- 位置：[章节/行号]
- 当前状态：[现有内容]
- 预期状态：[应有的内容]
- 修复方案：[具体修复措施]

ralph-prompt-single-task

[... same format]

[... 相同格式]

ralph-prompt-multi-task

[... same format]

[... 相同格式]

ralph-prompt-project

[... same format]

[... 相同格式]

ralph-prompt-research

[... same format]

[... 相同格式]

Recommendations Summary

建议摘要

Must Fix (Critical/High)

必须修复（严重/高优先级）

[Recommendation 1]
[Recommendation 2]

[建议1]
[建议2]

Should Fix (Medium)

应该修复（中优先级）

[Recommendation 3]

[建议3]

Nice to Have (Low)

可优化项（低优先级）

[Recommendation 4]

[建议4]

Review Result

审查结果

CLEAN (zero recommendations)
NEEDS_WORK (has recommendations)

---

CLEAN（无改进建议）
NEEDS_WORK（有改进建议）

---

STEP 3: IMPLEMENT IMPROVEMENTS

步骤3：实施改进

If REVIEW_FINDINGS.md shows NEEDS_WORK:

如果REVIEW_FINDINGS.md显示NEEDS_WORK：

3.1 Prioritize Fixes

3.1 优先处理修复任务

Work in this order:

Critical issues (breaks functionality)
High issues (significantly impacts quality)
Medium issues (improves quality)
Low issues (polish)

按以下顺序处理：

严重问题（影响功能运行）
高优先级问题（显著影响质量）
中优先级问题（提升质量）
低优先级问题（优化 polish）

3.2 Implement Each Fix

3.2 实施每项修复

For each recommendation:

Read the target skill file
Implement the specific fix
Verify the fix addresses the issue
Commit the change:

bash

git add .claude/skills/ralph-prompt-[name]/SKILL.md
git commit -m "Improve ralph-prompt-[name]: [brief description]

- [Change 1]
- [Change 2]

Part of Ralph skills self-improvement loop iteration [N]"

针对每个建议：

读取目标技能文件
实施具体修复
验证修复是否解决了问题
提交变更：

bash

git add .claude/skills/ralph-prompt-[name]/SKILL.md
git commit -m "Improve ralph-prompt-[name]: [简要描述]

- [变更1]
- [变更2]

Part of Ralph skills self-improvement loop iteration [N]"

3.3 Track Improvements

3.3 跟踪改进

Update RALPH_REVIEW_STATE.json:

json

{
  "improvements_made": [
    {
      "iteration": N,
      "skill": "ralph-prompt-X",
      "issue": "description",
      "fix": "what was done"
    }
  ]
}

更新RALPH_REVIEW_STATE.json：

json

{
  "improvements_made": [
    {
      "iteration": N,
      "skill": "ralph-prompt-X",
      "issue": "description",
      "fix": "what was done"
    }
  ]
}

STEP 4: POST-IMPROVEMENT VERIFICATION

步骤4：改进后验证

After implementing fixes:

实施修复后：

4.1 Verify Each Skill Still Works

4.1 验证每个技能仍可正常工作

For each modified skill:

YAML frontmatter is valid
All sections render correctly
Examples are syntactically correct
No broken references

针对每个修改过的技能：

YAML前置元数据有效
所有部分渲染正确
示例语法正确
无失效引用

4.2 Check for Regressions

4.2 检查是否出现回归

No content accidentally deleted
Cross-references still valid
Templates still complete

无内容被意外删除
交叉引用仍然有效
模板仍然完整

4.3 Run Syntax Check

4.3 运行语法检查

bash

undefined

bash

undefined

Verify YAML frontmatter

验证YAML前置元数据

for f in .claude/skills/ralph-prompt-*/SKILL.md; do head -20 "$f" | grep -E "^(name:|description:)" done

---

for f in .claude/skills/ralph-prompt-*/SKILL.md; do head -20 "$f" | grep -E "^(name:|description:)" done

---

STEP 5: UPDATE STATE

步骤5：更新状态

Update RALPH_REVIEW_STATE.json:

If review was CLEAN (zero recommendations):

json

{
  "consecutive_clean_reviews": [previous + 1],
  "last_review_result": "CLEAN",
  "last_review_timestamp": "[timestamp]"
}

If review was NEEDS_WORK:

json

{
  "consecutive_clean_reviews": 0,
  "last_review_result": "NEEDS_WORK",
  "improvements_this_iteration": [count],
  "last_review_timestamp": "[timestamp]"
}

Update RALPH_REVIEW_LOG.md:

markdown

undefined

更新RALPH_REVIEW_STATE.json：

如果审查结果为CLEAN（无改进建议）：

json

{
  "consecutive_clean_reviews": [上一次数值 + 1],
  "last_review_result": "CLEAN",
  "last_review_timestamp": "[时间戳]"
}

如果审查结果为NEEDS_WORK：

json

{
  "consecutive_clean_reviews": 0,
  "last_review_result": "NEEDS_WORK",
  "improvements_this_iteration": [数量],
  "last_review_timestamp": "[时间戳]"
}

更新RALPH_REVIEW_LOG.md：

markdown

undefined

Iteration [N] - [timestamp]

迭代[N] - [时间戳]

Review Result

审查结果

[CLEAN/NEEDS_WORK]

Issues Found

发现的问题

[Issue 1]
[Issue 2]

[问题1]
[问题2]

Fixes Applied

已应用的修复

[Fix 1]
[Fix 2]

[修复1]
[修复2]

State After

后续状态

Consecutive clean reviews: [N]
Total improvements to date: [N]

---

连续无问题审查次数：[N]
累计改进次数：[N]

---

STEP 6: LOOP DECISION

步骤6：循环决策

Check Termination Condition

检查终止条件

Read RALPH_REVIEW_STATE.json:

bash

cat RALPH_REVIEW_STATE.json | jq '.consecutive_clean_reviews'

读取RALPH_REVIEW_STATE.json：

bash

cat RALPH_REVIEW_STATE.json | jq '.consecutive_clean_reviews'

If consecutive_clean_reviews >= 2:

如果consecutive_clean_reviews >= 2：

Skills have passed two consecutive reviews with zero recommendations.

Create RALPH_SKILLS_VALIDATION_COMPLETE.md:

markdown

undefined

技能已连续两次通过审查且无任何改进建议。

创建RALPH_SKILLS_VALIDATION_COMPLETE.md：

markdown

undefined

Ralph Skills Validation Complete

Ralph技能验证完成

Summary

摘要

Total iterations: [N]
Total improvements made: [count]
Final state: All skills validated

总迭代次数：[N]
累计改进次数：[数量]
最终状态：所有技能已验证

Skills Validated

已验证技能

ralph-prompt-builder - PASSED
ralph-prompt-single-task - PASSED
ralph-prompt-multi-task - PASSED
ralph-prompt-project - PASSED
ralph-prompt-research - PASSED

ralph-prompt-builder - 通过
ralph-prompt-single-task - 通过
ralph-prompt-multi-task - 通过
ralph-prompt-project - 通过
ralph-prompt-research - 通过

Validation Criteria Met

满足的验证标准

All 16 review criteria passing for all 5 skills.

所有5项技能均通过全部16项审查标准。

Timestamp

时间戳

[ISO timestamp]


Output: <promise>RALPH_SKILLS_PERFECTED</promise>

[ISO时间戳]


输出：<promise>RALPH_SKILLS_PERFECTED</promise>

If consecutive_clean_reviews < 2:

如果consecutive_clean_reviews < 2：

Continue to next iteration (loop back to STEP 1)

继续下一次迭代（回到步骤1）

REVIEW CRITERIA REFERENCE (Quick Check)

审查标准参考（快速检查）

Ralph Technique Alignment

Ralph技术对齐度

Clear completion criteria
Self-verification commands
TDD/iteration approach
"If Stuck" guidance
<promise> tags used correctly
Appropriate max-iterations recommendations
Deterministically bad philosophy

清晰的完成标准
自我验证命令
TDD/迭代方法
"遇到问题时"的指导
<promise>标签正确使用
合适的max-iterations推荐
确定性可修复理念

Skill Structure Quality

技能结构质量

Complete YAML frontmatter
Progressive disclosure
Actionable Quick Start
Realistic examples
Best practices section
Ralph loop integration docs

完整的YAML前置元数据
渐进式披露
可操作的Quick Start
真实的示例
最佳实践部分
Ralph循环集成文档

Content Completeness

内容完整性

No placeholders
Matching examples
Real verification commands

无占位符
匹配的示例
真实的验证命令

ESCAPE HATCH

应急方案

If stuck after 50 iterations without reaching 2 consecutive clean reviews:

Document the recurring issues in RALPH_REVIEW_BLOCKERS.md
List which criteria keep failing
Identify if criteria are too strict
Output: <promise>RALPH_REVIEW_BLOCKED</promise>

如果经过50次迭代仍未达到2次连续无问题审查：

在RALPH_REVIEW_BLOCKERS.md中记录反复出现的问题
列出持续未通过的标准
判断标准是否过于严格
输出：<promise>RALPH_REVIEW_BLOCKED</promise>

PROGRESS TRACKING

进度跟踪

Every 5 iterations, summarize:

PROGRESS SUMMARY - Iteration [N]
================================
Started: [timestamp]
Current: [timestamp]
Consecutive clean reviews: [N]/2

Skills Status:
- ralph-prompt-builder: [X/16 criteria passing]
- ralph-prompt-single-task: [X/16 criteria passing]
- ralph-prompt-multi-task: [X/16 criteria passing]
- ralph-prompt-project: [X/16 criteria passing]
- ralph-prompt-research: [X/16 criteria passing]

Improvements made: [total count]
Remaining issues: [count]

每5次迭代，输出摘要：

PROGRESS SUMMARY - Iteration [N]
================================
开始时间：[时间戳]
当前时间：[时间戳]
连续无问题审查次数：[N]/2

技能状态：
- ralph-prompt-builder: [X/16项标准通过]
- ralph-prompt-single-task: [X/16项标准通过]
- ralph-prompt-multi-task: [X/16项标准通过]
- ralph-prompt-project: [X/16项标准通过]
- ralph-prompt-research: [X/16项标准通过]

已完成改进：[总数量]
剩余问题：[数量]

COMPLETION CONDITIONS

完成条件

Output <promise>RALPH_SKILLS_PERFECTED</promise> ONLY when:

All 5 skills reviewed
All 16 criteria checked per skill
Zero recommendations in current review
Zero recommendations in previous review
consecutive_clean_reviews >= 2 in state file
RALPH_SKILLS_VALIDATION_COMPLETE.md created
All changes committed

仅在满足以下所有条件时输出<promise>RALPH_SKILLS_PERFECTED</promise>：

所有5项技能已审查
每项技能均检查了全部16项标准
当前审查无任何建议
上一次审查无任何建议
状态文件中consecutive_clean_reviews >= 2
已创建RALPH_SKILLS_VALIDATION_COMPLETE.md
所有变更已提交

SAFETY LIMITS

安全限制

Maximum iterations: 100
Expected completion: 20-40 iterations
Budget alert: If > 50 iterations, evaluate if criteria are too strict

---

最大迭代次数：100
预期完成次数：20-40次迭代
预算预警：如果超过50次迭代，评估标准是否过于严格

---

Usage Instructions

使用说明

1. Initialize State Files

1. 初始化状态文件

Before running, create the initial state:

bash

undefined

运行前，创建初始状态：

bash

undefined

Create state file

创建状态文件

cat > RALPH_REVIEW_STATE.json << 'EOF' { "current_iteration": 0, "consecutive_clean_reviews": 0, "skills_reviewed": [], "improvements_made": [], "last_review_timestamp": "", "status": "NOT_STARTED" } EOF

Create log file

创建日志文件

cat > RALPH_REVIEW_LOG.md << 'EOF'

Ralph Skills Review Log

Ralph技能审查日志

Overview

概述

Self-improving review loop for Ralph Wiggum prompt generator skills.

针对Ralph Wiggum提示生成器技能的自我改进审查循环。

Target: Two consecutive clean reviews

目标：两次连续无问题审查

Iteration History

迭代历史

EOF

undefined

EOF

undefined

2. Run the Loop

2. 运行循环

bash

/ralph-wiggum:ralph-loop "[THE PROMPT ABOVE]" \
  --completion-promise "RALPH_SKILLS_PERFECTED" \
  --max-iterations 100

bash

/ralph-wiggum:ralph-loop "[THE PROMPT ABOVE]" \
  --completion-promise "RALPH_SKILLS_PERFECTED" \
  --max-iterations 100

3. Monitor Progress

3. 监控进度

bash

undefined

bash

undefined

Check current state

检查当前状态

cat RALPH_REVIEW_STATE.json | jq '.'

See recent activity

查看近期活动

tail -30 RALPH_REVIEW_LOG.md

Check how many clean reviews

检查无问题审查次数

cat RALPH_REVIEW_STATE.json | jq '.consecutive_clean_reviews'

undefined

cat RALPH_REVIEW_STATE.json | jq '.consecutive_clean_reviews'

undefined

4. After Completion

4. 完成后

Review the outputs:

```
RALPH_REVIEW_STATE.json
```
- Final state
```
RALPH_REVIEW_LOG.md
```
- Complete history
```
REVIEW_FINDINGS.md
```
- Last review details
```
RALPH_SKILLS_VALIDATION_COMPLETE.md
```
- Success certificate
Git log - All improvements committed

查看输出内容：

```
RALPH_REVIEW_STATE.json
```
- 最终状态
```
RALPH_REVIEW_LOG.md
```
- 完整历史
```
REVIEW_FINDINGS.md
```
- 最后一次审查详情
```
RALPH_SKILLS_VALIDATION_COMPLETE.md
```
- 成功验证证书
Git日志 - 所有改进的提交记录

Why This Works

为什么此方案有效

State Tracking: JSON state file persists across iterations
Clear Criteria: 16 specific, measurable review criteria
Self-Correction: Each iteration reads previous results and fixes issues
Termination Condition: Two consecutive clean reviews ensures stability
Evidence-Based: All findings documented, all fixes tracked
Git Integration: Every improvement committed for auditability

状态跟踪：JSON状态文件在迭代间持久化存储
清晰标准：16项具体、可衡量的审查标准
自我修正：每次迭代读取之前的结果并修复问题
终止条件：两次连续无问题审查确保稳定性
基于证据：所有发现均有文档记录，所有修复均被跟踪
Git集成：每项改进均已提交，便于审计

Expected Behavior

预期行为

Iteration 1-5: Discovery phase

Identify initial issues across all skills
Begin fixing critical issues

Iteration 6-15: Improvement phase

Systematic fixes
Quality improvements
Cross-consistency

Iteration 16-25: Stabilization phase

Fewer issues found
Polish and edge cases
Approaching clean reviews

Iteration 26-40: Validation phase

First clean review achieved
Verify no regressions
Second clean review achieved
Completion

迭代1-5：发现阶段

识别所有技能的初始问题
开始修复严重问题

迭代6-15：改进阶段

系统性修复问题
提升质量
确保交叉一致性

迭代16-25：稳定阶段

发现的问题减少
优化细节和边缘情况
接近无问题审查

迭代26-40：验证阶段

首次获得无问题审查
验证无回归问题
获得第二次无问题审查
完成循环

Customization

自定义配置

Stricter Review

更严格的审查

Add more criteria to the review framework.

在审查框架中添加更多标准。

Faster Completion

更快完成

Reduce to "one clean review" by changing:

consecutive_clean_reviews >= 1

将终止条件改为“一次无问题审查”：

consecutive_clean_reviews >= 1

Focus on Specific Skills

聚焦特定技能

Modify the target skills list in the prompt.

修改提示中的目标技能列表。

ralph-skill-review-loop

Original

Translation

Ralph Skill Review Loop

Ralph技能审查循环

Overview

概述

Quick Start

快速开始

THE SELF-IMPROVING REVIEW LOOP PROMPT

自我改进审查循环提示

Task: Self-Improving Review of Ralph Wiggum Skills

任务：Ralph Wiggum技能的自我改进审查

Objective

目标

Target Skills

目标技能

Reference Materials

参考资料

STATE MANAGEMENT

状态管理

Required State Files

所需状态文件

Ralph Skills Review Log

Ralph技能审查日志

Iteration History

迭代历史

STEP 1: ORIENTATION (Every Iteration)

步骤1：定位（每次迭代）

STEP 2: COMPREHENSIVE SKILL REVIEW

步骤2：全面技能审查

Review Framework

审查框架

2.1 Ralph Technique Alignment (from research)

2.1 Ralph技术对齐度（来自研究内容）

2.2 Skill Structure Quality

2.2 技能结构质量

2.3 Content Completeness

2.3 内容完整性

2.4 Prompt Template Quality

2.4 提示模板质量

Review Process

审查流程

Review Output Format

审查输出格式

Review Findings - Iteration [N]

审查发现 - 迭代[N]

Summary

摘要

ralph-prompt-builder

ralph-prompt-builder

Passing

通过项

Issues Found

发现的问题

ralph-prompt-single-task

ralph-prompt-single-task

ralph-prompt-multi-task

ralph-prompt-multi-task

ralph-prompt-project

ralph-prompt-project

ralph-prompt-research

ralph-prompt-research

Recommendations Summary

建议摘要

Must Fix (Critical/High)

必须修复（严重/高优先级）

Should Fix (Medium)

应该修复（中优先级）

Nice to Have (Low)

可优化项（低优先级）

Review Result

审查结果

STEP 3: IMPLEMENT IMPROVEMENTS

步骤3：实施改进

3.1 Prioritize Fixes

3.1 优先处理修复任务

3.2 Implement Each Fix

3.2 实施每项修复

3.3 Track Improvements