analyze-changes

Interactive chunk-by-chunk review of PR or branch changes. Identifies logical groupings, explains each chunk, and critiques for issues. Use when user says "analyze-changes", "review changes", "review this PR", "what changed", or "walk me through these changes".

16 stars

Best use case

analyze-changes is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Interactive chunk-by-chunk review of PR or branch changes. Identifies logical groupings, explains each chunk, and critiques for issues. Use when user says "analyze-changes", "review changes", "review this PR", "what changed", or "walk me through these changes".

Teams using analyze-changes should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/analyze-changes/SKILL.md --create-dirs "https://raw.githubusercontent.com/diegosouzapw/awesome-omni-skill/main/skills/backend/analyze-changes/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/analyze-changes/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How analyze-changes Compares

Feature / Agentanalyze-changesStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Interactive chunk-by-chunk review of PR or branch changes. Identifies logical groupings, explains each chunk, and critiques for issues. Use when user says "analyze-changes", "review changes", "review this PR", "what changed", or "walk me through these changes".

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Analyze Changes

Perform an interactive, chunk-by-chunk review of changes in a PR or branch. Explain what each logical change does, then critique for potential issues.

## When to Use

- Reviewing a PR before merge
- Understanding what changed on a feature branch
- Self-review before pushing
- Onboarding to someone else's changes

## Process

### 1. Detect Context

Determine the diff source:

**If on a PR branch or user mentions PR:**
```bash
gh pr view --json number,title,body,baseRefName 2>/dev/null
gh pr diff
```

**If on a feature branch (no PR):**
```bash
# Detect main branch
git symbolic-ref refs/remotes/origin/HEAD 2>/dev/null | sed 's@^refs/remotes/origin/@@'
# Or default to main/master

git diff main...HEAD
# or
git diff master...HEAD
```

**If user specifies:**
- Follow their instruction (specific commits, files, etc.)

### 2. Identify Logical Chunks

Analyze the diff and group changes into logical chunks. A chunk is:
- A coherent unit of change (single purpose)
- May span multiple files
- Should be reviewable independently

**Good chunking:**
- "Add user validation middleware"
- "Refactor database connection pooling"
- "Update API response format"
- "Add tests for auth flow"

**Bad chunking:**
- By file (misses logical grouping)
- Too granular (every function)
- Too broad ("all the changes")

### 3. Present Overview

```markdown
## Change Overview

**Context**: {PR #N: title | Branch: name vs main}

**Overall Goal**: {1-2 sentences: what does this PR/branch aim to achieve}

## Logical Chunks

1. **{Chunk 1 theme}**: {one-line description}
2. **{Chunk 2 theme}**: {one-line description}
3. **{Chunk 3 theme}**: {one-line description}

---

Ready to review Chunk 1: "{theme}"? (y/skip/all)
```

### 4. Review Each Chunk

Wait for user confirmation before each chunk.

**User responses:**
- `y` / `yes` / `continue` / `proceed` → Review this chunk
- `skip` → Skip to next chunk
- `all` → Review all remaining chunks without pausing

For each chunk, follow [CHUNK-REVIEW-GUIDE.md](CHUNK-REVIEW-GUIDE.md):

```markdown
### Chunk {N}: {Theme}

**Files touched**:
- `{path/to/file1}` — {brief: added/modified/deleted, what}
- `{path/to/file2}` — {brief}

**What this changes**:
{Explanation of what the code does now, how it differs from before}

**Why** (if discernible):
{Rationale for the change — from PR description, commit messages, or code context}

**Concerns**:
{Any issues, risks, or suggestions — or "None identified" if solid}

---

Ready for Chunk {N+1}: "{theme}"? (y/skip/all)
```

### 5. Final Summary

After all chunks reviewed:

```markdown
## Review Summary

**Chunks reviewed**: {N} of {total}

**Cross-cutting concerns**:
{Issues that span multiple chunks or affect the PR as a whole}

**Overall assessment**:
{Brief: is this PR ready? What needs attention?}
```

## Guidelines

- **Explain before critiquing**: User should understand the change before hearing concerns
- **Be specific in concerns**: "This might break X because Y" not "This seems risky"
- **Acknowledge good work**: If a chunk is well-done, say so briefly
- **Don't invent issues**: If code is solid, "None identified" is fine
- **Respect user's time**: Keep explanations concise, expand if asked

Related Skills

excel-field-analyzer

16
from diegosouzapw/awesome-omni-skill

分析Excel/CSV字段结构,AI自动生成中英文映射,验证翻译质量,输出统计报告。用于电子表格分析、数据字典创建、字段映射场景。

analyze-us-bank-credit-deposit-decoupling

16
from diegosouzapw/awesome-omni-skill

分析銀行貸款與存款之間的「信貸創造脫鉤」現象,追蹤存款的絕對收縮與回升軌跡,用以辨識聯準會緊縮政策在銀行體系內部的真實傳導效果。

analyze-silver-miner-metal-ratio

16
from diegosouzapw/awesome-omni-skill

以「銀礦股價格 ÷ 白銀價格」的相對比率衡量礦業股板塊相對於金屬本體的估值區間(偏貴/偏便宜),並用歷史分位數與類比區間推導「底部/頂部」訊號與情境推演。

analyze-jgb-insurer-superlong-flow

16
from diegosouzapw/awesome-omni-skill

從日本保險公司對超長期(10年以上)JGB 的淨買賣時間序列,自動產出「本月是否創紀錄淨賣出、連續淨賣出月數、期間累積淨賣出」等結論。

analyze-japan-debt-service-tax-burden

16
from diegosouzapw/awesome-omni-skill

以日本公債殖利率變化為觸發,量化「政府利息支出 / 稅收」負擔(含情境壓力測試),並判斷是否進入債務利息螺旋風險區。

analyze-investment-clock-rotation

16
from diegosouzapw/awesome-omni-skill

把「獲利成長 × 財務狀況(金融環境)」映射成「投資時鐘」,判斷目前落在哪個象限、近期是順時針還是逆時針旋轉、以及相對於上一輪循環的位置差異。

analyze-copper-supply-concentration-risk

16
from diegosouzapw/awesome-omni-skill

用公開資料量化「銅供應是否過度集中、主要產地是否結構性衰退、替代增量是否依賴少數國家」,並輸出可行的中期供應風險結論與情境推演。

analyze-copper-stock-resilience-dependency

16
from diegosouzapw/awesome-omni-skill

用跨資產訊號(全球股市韌性 + 中國利率環境)評估銅價能否突破關卡或進入「回補/回踩」到支撐的機率與路徑。

analyze-agent-overlap

16
from diegosouzapw/awesome-omni-skill

Analyzes existing agents, skills, prompts, and instructions to identify overlaps, redundancies, and conflicts. Works with GitHub Copilot, Claude Code, Codex, OpenCode, and other providers. Use before creating new customization files to avoid duplication, when consolidating agents, or when troubleshooting conflicting behaviors.

ab-testing-analyzer

16
from diegosouzapw/awesome-omni-skill

全面的AB测试分析工具,支持实验设计、统计检验、用户分群分析和可视化报告生成。用于分析产品改版、营销活动、功能优化等AB测试结果,提供统计显著性检验和深度洞察。

32-analyze-verify-150

16
from diegosouzapw/awesome-omni-skill

[32] ANALYZE. Ensure every critical claim has verifiable evidence with confidence levels. Each fact must have source + confidence percentage. If confidence <85%, enter Loop150 to find more sources. Use for critical decisions, factual claims, legal/compliance work, or any situation where unverified claims are dangerous.

video-analyzer

16
from diegosouzapw/awesome-omni-skill

鏅鸿兘鍒嗘瀽 Bilibili/YouTube/鏈湴瑙嗛锛岀敓鎴愯浆鍐欍€佽瘎浼板拰鎬荤粨銆傛敮鎸佸叧閿抚鎴浘鑷姩宓屽叆銆?