Code Review Assistant

An AI-powered code review system built with FastAPI and OpenAI, showcasing GenAI expertise, prompt engineering, and practical AI application development.

📁 Project Structure

code-review-assistant/
├── src/
│   ├── llm_handler.py          # OpenAI API integration
│   ├── prompt_templates.py     # Engineered prompts
│   └── utils.py                # Code validation utilities
├── test_prompts.py             # Test suite
├── requirements.txt            # Dependencies
├── .env.example               # Environment template
└── README.md                  # Documentation

1. Installation

# Clone the repository
git clone <your-repo-url>
cd code-review-assistant

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

2. Configuration

# Copy environment template
cp .env.example .env

# Edit .env and add your OpenAI API key
OPENAI_API_KEY=your_api_key_here

3. Test the Setup

python test_prompts.py

This will run several tests demonstrating:

• Code validation and metrics
• Comprehensive code review
• Unit test generation
• Different review types (security, performance, best practices)

Review Types Available

1. Comprehensive Review: Complete code analysis covering all aspects
2. Security Review: Focus on vulnerabilities and security issues
3. Performance Review: Analyze complexity and optimization opportunities
4. Best Practices: Check adherence to coding standards and patterns
5. Documentation Review: Evaluate comments and documentation quality
6. Bug Detection: Identify and fix potential bugs
7. Complexity Analysis: Big O notation and scalability assessment

💡 Example Usage

from src.llm_handler import get_openai_handler
from src.prompt_templates import PromptTemplates, ReviewType

# Initialize handler
handler = get_openai_handler()

# Review code
code = """
def calculate_sum(numbers):
    total = 0
    for num in numbers:
        total += num
    return total
"""

# Generate review
system_prompt = get_system_prompt(ReviewType.PERFORMANCE)
user_prompt = PromptTemplates.get_code_review_prompt(
    code=code,
    review_type=ReviewType.PERFORMANCE
)

response = await handler.generate_completion(
    prompt=user_prompt,
    system_prompt=system_prompt
)

print(response["content"])

Prompts

• Role-based system prompts for different expertise areas
• Structured output formatting
• Context-aware analysis
• Specific, actionable recommendations

Code Analysis Utilities

• Python syntax validation using AST
• Function and class extraction
• Line counting (code, comments, blank)
• Token estimation for context management

LLM Handler

• Async and sync completion methods
• Streaming support (for real-time responses)
• Usage tracking and logging
• Error handling

Phase 3: FastAPI Endpoints

• RESTful API for code review
• Request/response models with Pydantic
• File upload support
• Rate limiting

Phase 4: Advanced Features

• Multi-file analysis
• RAG for code pattern recognition
• Batch processing
• Review history and caching

Tech Stack

• Backend: FastAPI, Uvicorn
• AI/ML: OpenAI GPT-4
• Code Analysis: Python AST, Radon, Bandit
• Async: asyncio, aiofiles
• Config: python-dotenv

Environment Variables

Variable	Description	Default
OPENAI_API_KEY	OpenAI API key	Required
OPENAI_MODEL	Model to use	gpt-4-turbo-preview
TEMPERATURE	Sampling temperature	0.3
MAX_TOKENS	Max tokens per response	4000
MAX_CODE_LENGTH	Max code length to analyze	10000