code-review

name	code-review
description	当用户要求"审查代码"、"代码检查"、"检查bug"、"代码质量检查"、"发现问题"、"重构建议"，或者提到"代码审查"、"code review"、"审查"、"检查"时使用此技能。用于代码质量、潜在bug、性能问题、安全漏洞或改进机会的反馈。
version	2.0.0

Code Review Skill

Comprehensive code review guidance for Tauri + Rust + React applications with focus on quality, security, and maintainability.

Overview

This skill provides systematic code review guidance covering:

• Correctness: Logic errors, edge cases, type safety
• Performance: Inefficient algorithms, resource leaks, optimization opportunities
• Security: SQL injection, XSS, unsafe operations, data validation
• Maintainability: Code duplication, naming, documentation, test coverage
• Tauri-specific: Command patterns, specta attributes, desktop app constraints
• Project-specific: Adherence to CLAUDE.md guidelines

When This Skill Applies

This skill activates when:

• Reviewing pull requests or code changes
• Performing code quality checks
• Looking for bugs and issues
• Suggesting refactoring opportunities
• Auditing security vulnerabilities
• Checking performance bottlenecks
• Ensuring compliance with project standards

Code Review Framework

1. Pre-Review Checklist

Before diving into code details:

# Check git diff context
git diff main...feature-branch
git log --oneline main..feature-branch

# Verify tests pass
cargo test
npm test

# Check formatting
cargo fmt --check
npm run lint

Questions to answer:

• What is the purpose of this change?
• What files are modified? (Use git diff --stat)
• Are there breaking changes?
• Is test coverage adequate?
• Does it follow project conventions?

2. Rust Backend Review

Command Pattern Review

✅ Correct Tauri Command:

#[tauri::command]
#[specta::specta]
pub async fn get_materials(
    state: State<'_, TauriAppState>,
) -> ApiResponse<Vec<Material>> {
    with_service(state, |ctx| async move {
        ctx.material_service.get_all().await
    })
    .await
}

Checklist:

• #[tauri::command] attribute present
• #[specta::specta] attribute present
• Return type is ApiResponse<T>
• Uses with_service helper or proper error handling
• Async functions properly use await

Error Handling Review

✅ Good Error Handling:

pub async fn create_formula(&self, dto: CreateFormulaDto) -> Result<Formula> {
    // Validate input
    if dto.name.is_empty() {
        return Err(anyhow!("名称不能为空"));
    }

    // Check for duplicates
    if self.exists_by_name(&dto.name).await? {
        return Err(anyhow!("配方名称已存在"));
    }

    // Create with transaction
    let mut tx = self.pool.begin().await?;
    let id = self.insert_internal(&dto, &mut tx).await?;
    tx.commit().await?;

    Ok(self.get_by_id(id).await?)
}

Common Issues to Watch For:

• ❌ Missing error propagation (? not used)
• ❌ Swallowing errors (let _ = result)
• ❌ Generic error messages without context
• ❌ Not using .context() for error chaining
• ❌ Forgetting to commit transactions

Database Query Review

✅ Safe SQLx Query:

sqlx::query_as!(
    Material,
    "SELECT code, name, price FROM materials WHERE code = ?",
    code
)
.fetch_one(&pool)
.await

Common Issues:

• ❌ SELECT * - specify columns explicitly
• ❌ N+1 queries in loops
• ❌ String concatenation in SQL (injection risk)
• ❌ Missing parameter binding
• ❌ No indexes on frequently queried fields

Performance Review

Red Flags:

// ❌ Blocking async runtime
std::thread::sleep(Duration::from_secs(10));

// ❌ Unnecessary cloning
let data = self.large_data.clone(); // Can use &LargeData

// ❌ Inefficient data structures
let mut items = Vec::new();
for item in huge_list {
    if items.contains(&item) { // O(n) lookup
        items.push(item);
    }
}

// ✅ Use HashSet for O(1) lookups
use std::collections::HashSet;
let mut items = HashSet::new();

Performance Checklist:

• Appropriate use of Arc vs & references
• Efficient data structures (HashMap, HashSet vs Vec)
• Parallel processing with rayon for CPU-bound tasks
• Caching for frequently accessed data
• No blocking calls in async context

3. React Frontend Review

Component Quality Review

✅ Well-Structured Component:

import React, { useCallback, useMemo } from 'react';
import { message } from 'antd';

interface Props {
  formula: Formula;
  onUpdate: (id: number) => void;
}

export const FormulaCard: React.FC<Props> = React.memo(({ formula, onUpdate }) => {
  const totalCost = useMemo(
    () => formula.materials.reduce((sum, m) => sum + m.cost, 0),
    [formula.materials]
  );

  const handleUpdate = useCallback(() => {
    onUpdate(formula.id);
  }, [formula.id, onUpdate]);

  return (
    <div>
      <h3>{formula.name}</h3>
      <p>成本: {totalCost.toFixed(2)}</p>
      <Button onClick={handleUpdate}>更新</Button>
    </div>
  );
});

Checklist:

• Proper TypeScript types (no any)
• React.memo for performance
• useCallback for callbacks passed as props
• useMemo for expensive calculations
• No console.log (use message instead)
• Meaningful component names

Hooks Usage Review

✅ Correct Hooks Usage:

export const useFormulas = () => {
  const { data, isLoading, error } = useQuery({
    queryKey: ['formulas'],
    queryFn: () => commands.getFormulas(),
  });

  const createMutation = useMutation({
    mutationFn: (dto: CreateDto) => commands.createFormula(dto),
    onSuccess: () => {
      queryClient.invalidateQueries({ queryKey: ['formulas'] });
      message.success('创建成功');
    },
  });

  return { formulas: data, isLoading, createMutation };
};

Common Issues:

• ❌ Hooks called inside conditions/loops
• ❌ Missing dependencies in useEffect/useCallback
• ❌ State mutations instead of setState
• ❌ Using raw invoke instead of generated commands

4. Integration Review

Frontend-Backend Integration

Type Safety Check:

// ✅ Using generated types
import type { Formula, Material } from '../bindings';
import { commands } from '../bindings';

const result = await commands.getFormula(123);
if (!result.success) {
    message.error(result.message);
    return;
}
const formula: Formula = result.data; // Type-safe!

Issues to Watch:

• ❌ Using as any type assertions
• ❌ Not checking result.success
• ❌ Missing error handling
• ❌ Inconsistent type naming (snake_case vs camelCase)

5. Security Review

Common Vulnerabilities

SQL Injection:

// ❌ VULNERABLE
let query = format!("SELECT * FROM materials WHERE name = '{}'", name);
sqlx::query(&query).fetch_one(&pool).await

// ✅ SAFE
sqlx::query_as!(
    Material,
    "SELECT * FROM materials WHERE name = ?",
    name
)
.fetch_one(&pool).await

Input Validation:

// ❌ No validation
pub fn create_material(name: String, price: f64) -> Result<Material> {
    // Direct insert without checks
}

// ✅ With validation
pub fn create_material(name: String, price: f64) -> Result<Material> {
    if name.is_empty() || name.len() > 100 {
        return Err(anyhow!("无效的原料名称"));
    }
    if price < 0.0 || price > 10000.0 {
        return Err(anyhow!("价格超出合理范围"));
    }
    // Proceed with creation
}

Desktop App Security:

• No hardcoded credentials
• Environment variables for secrets
• File path validation (prevent directory traversal)
• Safe handling of user-provided file paths

6. Testing Review

Test Coverage Check

// ✅ Good test coverage
#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_calculate_cost() {
        // Normal case
        assert_eq!(calculate_cost(&materials), 100.0);

        // Edge cases
        assert_eq!(calculate_cost(&[]), 0.0);
        assert_eq!(calculate_cost(&[single_material]), single_material.cost);
    }

    #[test]
    fn test_negative_proportion_rejected() {
        let result = validate_proportion(-10.0);
        assert!(result.is_err());
    }
}

Checklist:

• Unit tests for business logic
• Edge cases covered (empty, null, negative values)
• Integration tests for database operations
• Error paths tested
• Mocking external dependencies

7. Documentation Review

What to Check:

• Public APIs have documentation comments
• Complex algorithms have explanations
• Non-obvious code has comments
• TODO/FIXME comments have associated issues
• README updated if behavior changed

Code Review Comment Templates

High Priority Issues

## 🔴 Critical: [Issue Title]

**Location**: `src/file.rs:42`

**Problem**: [Clear description of the issue]

**Impact**: [Why this matters]

**Suggested Fix**:
```rust
// Show corrected code

### Medium Priority Issues

```markdown
## 🟡 Suggestion: [Title]

**Location**: `src/file.rs:123`

**Current Approach**: [What the code does now]

**Recommendation**: [Better approach]

**Benefits**:
- [Benefit 1]
- [Benefit 2]

Minor Issues

## 💡 Nitpick: [Title]

**Location**: `src/file.rs:200`

**Observation**: [Small improvement]

**Why it matters**: [Optional explanation]

Common Issues by Category

Rust Issues

Issue	Pattern	Fix
Missing unwrap context	.unwrap()	.expect("Descriptive message") or proper error handling
Cloning instead of borrowing	.clone()	Use &T reference
Blocking async runtime	std::thread::sleep	tokio::time::sleep
SQL injection risk	format!("WHERE = {}", val)	Use ? parameter binding
N+1 query	Loop with query inside	Use JOIN or batch query

React Issues

Issue	Pattern	Fix
Console logging	console.log()	Use message component
Type assertion	as any	Use proper types
Missing deps	useEffect(fn, [])	Add all dependencies
Hook in condition	if (condition) { useState() }	Move to top level
Key prop issue	key={index}	Use unique ID

Review Workflow

1. Automated Checks (Do First)

# Run all automated checks
cargo test --all
cargo clippy -- -D warnings
cargo fmt --check
npm test
npm run lint
npm run type-check

2. Manual Review

1. Read diff top-to-bottom: Get overall picture
2. Check critical paths: Security, performance, correctness
3. Verify tests: Ensure adequate coverage
4. Check documentation: Is it up to date?
5. Test manually: If applicable, run the application

3. Provide Feedback

• Be constructive: Focus on improvement, not criticism
• Explain why: Help author understand the issue
• Provide examples: Show how to fix
• Prioritize: Mark critical vs. minor issues
• Be polite: Remember code reviews are learning opportunities

Quick Reference

Essential Commands

# Review changes
git diff main...feature-branch

# View specific file changes
git diff main..feature-branch -- src/file.rs

# Check commit history
git log --oneline main..feature-branch

# Run tests
cargo test
npm test

# Check formatting
cargo fmt --check
npm run lint

Review Checklist

Backend (Rust):
- [ ] Tauri command attributes correct
- [ ] Error handling comprehensive
- [ ] SQL queries safe and optimized
- [ ] Performance considerations addressed
- [ ] No blocking calls in async context

Frontend (React):
- [ ] TypeScript types correct
- [ ] Hooks properly used
- [ ] No console.log
- [ ] Error handling with message
- [ ] Performance optimized

Integration:
- [ ] Type-safe command calls
- [ ] Proper error handling
- [ ] Consistent naming
- [ ] specta types generated

Testing:
- [ ] Unit tests present
- [ ] Edge cases covered
- [ ] Integration tests if needed
- [ ] Tests passing

Documentation:
- [ ] Public APIs documented
- [ ] Complex logic explained
- [ ] README updated

When to Use This Skill

Activate this skill when:

• Reviewing pull requests
• Performing code quality audits
• Looking for bugs and issues
• Suggesting refactoring
• Checking for security vulnerabilities
• Ensuring project standard compliance
• Onboarding developers to codebase