🤖 AI Code Review API Documentation

Overview

The AI Code Review feature provides intelligent code analysis using Google Gemini (FREE) or OpenAI APIs. It supports real-time collaborative reviews with comments and sharing.

🚀 Quick Setup

1. Get FREE API Key

Visit Google AI Studio
Sign in with Google account
Click "Create API Key"
Copy the generated key

2. Environment Configuration

# FREE Gemini API (Recommended)
GEMINI_API_KEY=your_gemini_api_key_here
AI_PROVIDER=gemini

# Alternative: Paid OpenAI API
OPENAI_API_KEY=your_openai_api_key_here
AI_PROVIDER=openai

# Redis (Required for enhanced features)
REDIS_HOST=localhost
REDIS_PORT=6379

3. Start the Application

# Install dependencies
npm install

# Start Redis (if not running)
docker run -d --name redis -p 6379:6379 redis:alpine

# Start the application
npm start

📡 API Endpoints

1. Create AI Review

POST /api/ai-review/create

Creates a new collaborative AI code review.

Request Body:

{
  "roomId": "room-123",
  "code": "function hello() { console.log('Hello World'); }",
  "language": "javascript",
  "analysisTypes": ["bugs", "security", "performance", "style", "maintainability"]
}

Response:

{
  "success": true,
  "data": {
    "reviewId": "review_1234567890_abc123def",
    "summary": "Overall good code quality with minor style improvements needed",
    "overallScore": 85,
    "issueCount": 2,
    "analysis": {
      "overall_score": 85,
      "summary": "Well-structured function with clear intent",
      "issues": [...],
      "positive_aspects": [...],
      "recommendations": [...],
      "complexity_analysis": {...}
    }
  }
}

Real-time Event Emitted:

socket.emit('ai_review_created', {
  reviewId: "review_1234567890_abc123def",
  authorId: "user123",
  summary: "Overall good code quality...",
  overallScore: 85,
  issueCount: 2,
  language: "javascript",
  created_at: "2024-01-15T10:30:00.000Z"
});

2. Get Specific Review

GET /api/ai-review/:reviewId

Retrieves a specific review with all comments.

Response:

{
  "success": true,
  "data": {
    "id": "review_1234567890_abc123def",
    "roomId": "room-123",
    "authorId": "user123",
    "code": "function hello() { ... }",
    "language": "javascript",
    "analysis": {...},
    "created_at": "2024-01-15T10:30:00.000Z",
    "updated_at": "2024-01-15T10:35:00.000Z",
    "comments": [
      {
        "id": "comment_1234567890_xyz789",
        "reviewId": "review_1234567890_abc123def",
        "userId": "user456",
        "comment": "Great function structure!",
        "lineNumber": 1,
        "timestamp": "2024-01-15T10:32:00.000Z"
      }
    ]
  }
}

3. Get Room Reviews

GET /api/ai-review/room/:roomId?limit=10

Retrieves all reviews for a specific room.

Query Parameters:

limit (optional): Maximum number of reviews to return (default: 10)

Response:

{
  "success": true,
  "data": [
    {
      "id": "review_1234567890_abc123def",
      "authorId": "user123",
      "summary": "Well-structured function...",
      "overallScore": 85,
      "issueCount": 2,
      "language": "javascript",
      "created_at": "2024-01-15T10:30:00.000Z",
      "commentCount": 3
    }
  ]
}

4. Add Comment to Review

POST /api/ai-review/:reviewId/comment

Adds a comment to a specific review.

Request Body:

{
  "comment": "This function could be optimized further",
  "lineNumber": 5
}

Response:

{
  "success": true,
  "data": {
    "id": "comment_1234567890_xyz789",
    "reviewId": "review_1234567890_abc123def",
    "userId": "user456",
    "comment": "This function could be optimized further",
    "lineNumber": 5,
    "timestamp": "2024-01-15T10:35:00.000Z"
  }
}

Real-time Event Emitted:

socket.emit('review_comment_added', {
  reviewId: "review_1234567890_abc123def",
  comment: {
    id: "comment_1234567890_xyz789",
    userId: "user456",
    comment: "This function could be optimized further",
    lineNumber: 5,
    timestamp: "2024-01-15T10:35:00.000Z"
  },
  authorId: "user456"
});

5. Quick Code Analysis

POST /api/ai-review/analyze

Performs quick code analysis without storing results.

Request Body:

{
  "code": "const x = 5; console.log(x);",
  "language": "javascript",
  "analysisTypes": ["bugs", "style"]
}

Response:

{
  "success": true,
  "data": {
    "overall_score": 90,
    "summary": "Simple variable declaration and logging",
    "issues": [],
    "positive_aspects": ["Clear variable naming", "Simple logic"],
    "recommendations": ["Consider using more descriptive variable names"],
    "complexity_analysis": {
      "cyclomatic_complexity": "1",
      "maintainability_index": "90",
      "technical_debt": "low"
    },
    "metadata": {
      "language": "javascript",
      "analyzed_at": "2024-01-15T10:30:00.000Z",
      "code_length": 29,
      "line_count": 1
    }
  }
}

6. Get Service Status

GET /api/ai-review/status

Returns the current status of the AI service.

Response:

{
  "success": true,
  "data": {
    "enabled": true,
    "provider": "gemini",
    "isFree": true,
    "cacheSize": 25,
    "activeUsers": 5,
    "totalRequests": 150,
    "collaborativeReviews": 12,
    "activeRooms": 3,
    "totalComments": 45
  }
}

7. Batch Analysis (Authenticated Users Only)

POST /api/ai-review/batch-analyze

Analyzes multiple code snippets in a single request.

Request Body:

{
  "codeSnippets": [
    {
      "id": "snippet1",
      "code": "function add(a, b) { return a + b; }",
      "language": "javascript",
      "analysisTypes": ["bugs", "style"]
    },
    {
      "id": "snippet2", 
      "code": "def multiply(x, y): return x * y",
      "language": "python",
      "analysisTypes": ["style", "maintainability"]
    }
  ]
}

Response:

{
  "success": true,
  "data": [
    {
      "id": "snippet1",
      "analysis": {
        "summary": "Well-structured function",
        "overallScore": 95,
        "issueCount": 0,
        "criticalIssues": 0
      }
    },
    {
      "id": "snippet2",
      "analysis": {
        "summary": "Simple function with good structure", 
        "overallScore": 90,
        "issueCount": 1,
        "criticalIssues": 0
      }
    }
  ]
}

🎯 Analysis Types

Type	Description	Examples
`bugs`	Code errors and potential bugs	Null pointer exceptions, logic errors
`security`	Security vulnerabilities	SQL injection, XSS, unsafe operations
`performance`	Performance improvements	Algorithm efficiency, memory usage
`style`	Code style and formatting	Naming conventions, indentation
`maintainability`	Code maintainability	Complexity, readability, modularity
`complexity`	Code complexity analysis	Cyclomatic complexity, nesting depth
`documentation`	Documentation quality	Comments, function descriptions

🔒 Rate Limits

Provider	Requests per Minute	Daily Limit	Cost
Gemini (FREE)	15	1,500	FREE
OpenAI (PAID)	10	Unlimited*	~$0.0001-0.0005 per request

*Subject to your OpenAI billing limits

🌐 Real-time Events

The AI Review feature emits the following Socket.io events:

`ai_review_created`

Emitted when a new review is created in a room.

socket.on('ai_review_created', (data) => {
  console.log('New review:', data.reviewId);
  // Refresh reviews list
});

`review_comment_added`

Emitted when a comment is added to a review.

socket.on('review_comment_added', (data) => {
  console.log('New comment on review:', data.reviewId);
  // Update review display
});

🛠️ Error Handling

Common Error Responses

Rate Limit Exceeded (429):

{
  "success": false,
  "message": "Rate limit exceeded for Gemini (FREE). Please wait before requesting another review."
}

Service Unavailable (503):

{
  "success": false, 
  "message": "AI Code Review service is currently unavailable"
}

Code Too Large (400):

{
  "success": false,
  "message": "Code too large. Maximum 50,000 characters allowed."
}

Review Not Found (404):

{
  "success": false,
  "message": "Review not found"
}

🧪 Testing

Run the comprehensive test suite:

# Full test suite
node test-ai-review.js

# Quick tests only
node test-ai-review.js --quick

# Check test help
node test-ai-review.js --help

📈 Monitoring

The service includes automatic cleanup and monitoring:

Reviews older than 24 hours are automatically cleaned up
Rate limiting prevents abuse
Comprehensive metrics available via /api/ai-review/status
Memory usage is monitored and optimized

💡 Best Practices

Code Size: Keep code snippets under 10KB for optimal performance
Analysis Types: Select specific analysis types for faster results
Caching: Identical code analysis results are cached for 1 hour
Real-time: Use Socket.io events for collaborative features
Error Handling: Always handle rate limiting and service unavailability

🎯 Your AI-powered collaborative code review system is now fully functional! 🚀

FilesExpand file tree

AI-CODE-REVIEW-SETUP.md

Latest commit

History

AI-CODE-REVIEW-SETUP.md

File metadata and controls

🤖 AI Code Review API Documentation

Overview

🚀 Quick Setup

1. Get FREE API Key

2. Environment Configuration

3. Start the Application

📡 API Endpoints

1. Create AI Review

2. Get Specific Review

3. Get Room Reviews

4. Add Comment to Review

5. Quick Code Analysis

6. Get Service Status

7. Batch Analysis (Authenticated Users Only)

🎯 Analysis Types

🔒 Rate Limits

🌐 Real-time Events

ai_review_created

review_comment_added

🛠️ Error Handling

Common Error Responses

🧪 Testing

📈 Monitoring

💡 Best Practices

`ai_review_created`

`review_comment_added`