Skip to content

🚀 AI Code Reviewer Plus

AI-powered intelligent code reviewer — multi-dimensional semantic analysis, five-tier severity classification, actionable fix suggestions via Claude Code Skill.

npm versionlicense

The Problem

Traditional linters catch syntax errors, but miss semantic issues:

typescript
// Linter: ✅ No syntax errors
function processUser(input: any) {
  return input.name.trim()  // Runtime: 💥 if input is null
}

Questions linters can't answer:

  • Is this business logic correct?
  • Are there security vulnerabilities?
  • Will this cause performance issues?
  • Is this maintainable?

The Solution

AI Code Reviewer uses semantic understanding across five dimensions:

DimensionFocusExample Rules
CorrectnessLogic errors, null pointersCOR-001: Null check missing
SecurityXSS, SQL injection, data leakageSEC-001: User input unsanitized
PerformanceN+1 queries, memory leaksPER-001: Loop inside loop
MaintainabilityDuplicate code, complexityMAIN-001: Duplicated logic
Best PracticesFramework conventionsBP-001: Vue reactive rules

Five-Tier Severity

LevelMeaningAction
🚫 BLOCKERMust fix, blocks mergeFix immediately
🔴 HIGHCritical, breaks functionalityFix this iteration
🟡 MEDIUMImportant, affects qualityPlan to fix
🟢 LOWMinor, polish levelFix later
💡 SUGGESTIONOptional improvementOptional

Get Started

bash
# Claude Code Plugin (Recommended)
/plugin marketplace add saqqdy/ai-code-reviewer-plus
/plugin install ai-code-reviewer-plus

# CLI (Zero-Install)
npx ai-code-reviewer-plus review --branch main

# NPM Package
pnpm add ai-code-reviewer-plus

Project Status

VersionThemeStatus
v0.1.0Core framework + CLI + Skill✅ Current
v0.2.0Rule library + AI analysis engine📋 Planned
v1.0.0Production-ready + Marketplace📋 Planned

See Roadmap for details.

License

MIT — use freely in personal and commercial projects.

MIT License