AI Code Reviewer

An intelligent code review tool that leverages AI models to analyze code files and provide detailed feedback, suggestions, and improvements. The tool integrates with GitHub APIs to review pull requests and can analyze local code files.

Features

• AI-Powered Code Reviews: Uses Ollama with Llama3 model for intelligent code analysis
• Multiple Language Support: Supports Python, TypeScript, JavaScript, and JSX files
• GitHub Integration: Can analyze GitHub pull requests and commits
• Batch Processing: Review entire directories with exclude filters
• Markdown Reports: Generates detailed review reports in markdown format
• Customizable Prompts: Structured prompt templates for consistent review quality

Prerequisites

• Python 3.11+
• Ollama installed and running locally
• Llama3 model pulled in Ollama (ollama pull llama3)
• GitHub Personal Access Token (for GitHub integration)

Installation

1. Clone the repository:

git clone https://github.com/LiteObject/ai-code-reviewer.git
cd ai-code-reviewer

2. Install dependencies:

pip install -r requirements.txt

3. Set up environment variables:

# Create .env file
echo "GITHUB_TOKEN=your_github_token_here" > .env

4. Ensure Ollama is running:

ollama serve

Usage

Review Local Files

from reviewer import review_file, review_files

# Review a single file
review_file("path/to/your/file.py")

# Review all files in a directory
exclude_keywords = ["venv", "node_modules", "__pycache__"]
review_files("path/to/directory", exclude_keywords)

GitHub Integration

import git_client as git

# Get pull request data
pr_data = git.get_pull_request_data(token, owner, repo, pull_number)

# Get pull request changes
pr_changes = git.get_pull_request_changes(token, owner, repo, pull_number)

Run the Main Application

python app.py

Project Structure

ai-code-reviewer/
├── app.py                 # Main application entry point
├── reviewer.py            # Core review logic using LangChain
├── git_client.py          # GitHub API integration
├── file_helper.py         # File operations utilities
├── test_git_client.py     # Manual testing script
├── requirements.txt       # Python dependencies
├── .env                   # Environment variables (create this)
├── reviews/              # Generated review reports
│   ├── REVIEW_*_BY_llama3.md
└── test_files/           # Sample files for testing
    ├── bad_code_examples.py
    └── delete_folders.py

Configuration

Supported File Extensions

• .py (Python)
• .ts (TypeScript)
• .js (JavaScript)
• .jsx (React JSX)

Exclude Keywords

Configure which directories/files to skip during batch processing:

exclude_keywords = ["venv", "Python311", "node_modules", "__pycache__"]

Ollama Configuration

The tool uses Ollama running on http://localhost:11434 with the llama3 model. You can modify these settings in reviewer.py:

llm = Ollama(
    base_url='http://localhost:11434',
    model="llama3"
)

Review Output

Reviews are saved in the reviews/ directory with the format:

REVIEW_{filename}_BY_{model}.md

Each review includes:

• Code analysis summary
• Specific recommendations with before/after code examples
• Complete revised code when necessary
• Markdown formatting for easy reading

API Functions

Core Review Functions

• review_file(file_path) - Review a single file
• review_files(folder_path, exclude_keywords) - Review multiple files
• get_code_review(filename, code) - Generate AI review for code

GitHub Integration

• get_pull_request_data(token, owner, repo, pull_number)
• get_pull_request_changes(token, owner, repo, pull_number)
• get_commit_messages(token, owner, repo, pull_number)

File Operations

• create_file(filename, content) - Create new file
• append_to_file(filename, content) - Append to existing file

Dependencies

Key dependencies include:

• langchain - AI model integration
• requests - HTTP requests for GitHub API
• python-dotenv - Environment variable management
• pydantic - Data validation

See requirements.txt for complete list.

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

License

This project is open source. Please check the repository for license details.