Bug-Fixing-AI-Assistant 🤖🔧

An intelligent AI-powered assistant that automatically scans codebases for bugs, generates fixes using Claude or GPT, validates changes with automated tests, and integrates fixes into your codebase via pull requests.

Project Structure

Bug-Fixing-AI-Assistant/
├── scanner/                 # Code scanning and bug detection
│   ├── core/               # Core scanning functionality
│   │   ├── scanner.py      # Base scanner class
│   │   └── report.py       # Report generation
│   └── detectors/          # Bug detection modules
│       ├── pattern_detector.py    # Pattern-based bug detection
│       └── security_detector.py   # Security vulnerability detection
├── fixer/                  # Bug fixing and code generation
│   ├── generators/         # Fix generation modules
│   │   ├── fix_generator.py      # Base fix generator
│   │   └── patch_generator.py    # Patch and diff generation
│   └── validators/         # Fix validation modules
│       ├── fix_validator.py      # Fix syntax validation
│       └── test_runner.py        # Test execution
├── pr-handler/             # Pull request management
│   └── pr_creator.py       # PR creation and management
├── examples/               # Example buggy and fixed code
│   ├── buggy_code.py       # Intentionally buggy code samples
│   ├── fixed_code.py       # Fixed versions of buggy code
│   └── README.md           # Example documentation
└── tests/                  # Comprehensive test suite
    ├── test_scanner.py     # Scanner tests
    ├── test_detectors.py   # Detector tests
    ├── test_fixer.py       # Fixer tests
    ├── test_pr_handler.py  # PR handler tests
    └── README.md           # Test documentation

✨ Features

1. 🔍 Intelligent Code Scanning

Pattern Detection: Identifies common code anti-patterns
- None comparison using == instead of is
- Bare except clauses
- Mutable default arguments
- Wildcard imports
Security Analysis: Detects security vulnerabilities
- Dangerous eval() and exec() usage
- Unsafe deserialization (pickle)
- Insecure module imports
Code Quality: Analyzes code quality issues
- High cyclomatic complexity
- Too many function arguments
- Missing docstrings
- Magic numbers
- Functions with too many methods (God objects)

2. 🤖 AI-Powered Fix Generation

Claude Integration: Uses Anthropic's Claude for intelligent fixes
OpenAI Integration: Supports GPT-4 and GPT-3.5-turbo
Context-Aware: Provides surrounding code context to AI
Confidence Scoring: AI rates fix confidence level
Fallback Support: Falls back to rule-based fixes when AI unavailable

3. ⚡ Automated Fix Application

Smart Application: Automatically applies fixes to files
Backup Creation: Creates backups before modifying files
Syntax Validation: Validates fixed code syntax before applying
Dry Run Mode: Preview changes without modifying files
Batch Processing: Apply multiple fixes at once

4. ✅ Validation & Testing

Python Syntax Validation: Ensures fixes don't introduce syntax errors
Test Execution: Runs pytest or unittest after fixing
Configurable: Control whether tests must pass
Fix Verification: Validates fixes before committing

5. 🔄 Pull Request Integration

Automated Branching: Creates feature branches automatically
Smart Commits: Generates descriptive commit messages
PR Descriptions: Creates detailed PR descriptions with fix summaries
Auto-Push: Optionally pushes to remote automatically
GitHub Integration: Works with gh CLI for PR creation

🚀 Quick Start

Installation

Clone the repository:

git clone https://github.com/matthewnyc2/Bug-Fixing-AI-Assistant.git
cd Bug-Fixing-AI-Assistant

Install dependencies (optional, for AI features):

# For Anthropic Claude
pip install anthropic

# For OpenAI GPT
pip install openai

# Or install all optional dependencies
pip install -e ".[full]"

Set up your AI API key (optional):

# For Anthropic Claude
export ANTHROPIC_API_KEY=your-api-key-here

# For OpenAI GPT
export OPENAI_API_KEY=your-api-key-here

Command Line Usage

Basic Scanning (No AI)

# Scan current directory and show report
python main.py

# Scan a specific directory
python main.py /path/to/your/project

# Generate JSON report
python main.py --report json

AI-Powered Fixing

# Scan and generate AI fixes (dry run - no changes)
python main.py --ai --dry-run

# Scan, generate AI fixes, and apply them
python main.py --ai --apply

# Full workflow: scan, fix, test, and create PR
python main.py --ai --apply --run-tests --create-pr

Using Configuration File

# Copy example config
cp config.example.yaml config.yaml

# Edit config.yaml with your preferences
# Then run with config
python main.py --config config.yaml --ai --apply

Python API Usage

Simple Scanning

from main import BugFixingAssistant

# Create assistant
assistant = BugFixingAssistant()

# Scan directory
issues = assistant.scan_directory('.')

# Print issues
for issue in issues:
    print(f"{issue['type']} at {issue['file']}:{issue['line']}")

AI-Powered Fixing

from main import BugFixingAssistant
from config import Config

# Create config with AI enabled
config = Config()
config.set('ai.provider', 'anthropic')
config.set('ai.model', 'claude-sonnet-4-5-20250929')

# Create assistant
assistant = BugFixingAssistant(config)

# Scan and fix
issues = assistant.scan_directory('.')
fixes = assistant.generate_fixes(use_ai=True)

# Apply fixes (with backup)
results = assistant.apply_fixes()

# Show what was fixed
for result in results:
    if result['success']:
        print(f"✓ Fixed {result['file']}")
        print(result['diff'])

Create Pull Request

from main import BugFixingAssistant

assistant = BugFixingAssistant()
issues = assistant.scan_directory('.')
fixes = assistant.generate_fixes()
assistant.apply_fixes()

# Create PR with fixes
pr_result = assistant.create_pr(fixes)
print(pr_result['pr_description'])

Running Tests

Run all tests:

python -m unittest discover tests -v

Run specific test suite:

python -m unittest tests.test_scanner -v
python -m unittest tests.test_detectors -v
python -m unittest tests.test_fixer -v
python -m unittest tests.test_pr_handler -v

Examples

The examples/ directory contains sample files demonstrating the assistant's capabilities:

buggy_code.py: Contains intentional bugs for testing
fixed_code.py: Shows corrected versions of the bugs
README.md: Detailed explanation of each bug and fix

Test the scanner on examples:

python -c "
from scanner.detectors.pattern_detector import detect_patterns
from scanner.detectors.security_detector import detect_security_issues
from pathlib import Path

file_path = Path('examples/buggy_code.py')
with open(file_path) as f:
    content = f.read()

issues = detect_patterns(file_path, content) + detect_security_issues(file_path, content)
print(f'Found {len(issues)} issues')
"

Development

Project Requirements

Python 3.7+
No external dependencies for core functionality
Uses Python standard library only

Contributing

Fork the repository
Create a feature branch
Make your changes
Run tests to ensure everything works
Submit a pull request

Architecture

Scanner Module

The scanner module is responsible for analyzing code and detecting potential issues. It uses Python's ast module for syntax tree analysis.

Fixer Module

The fixer module generates suggested fixes for detected issues. It includes validators to ensure fixes don't introduce new problems.

PR Handler Module

The PR handler automates the process of creating pull requests with fixes, including branch management and commit operations.

📋 Supported Issue Types

Issue Type	Severity	Auto-Fix	AI-Fix
Pattern Issues
None comparison (`==` instead of `is`)	Warning	✅ Yes	✅ Yes
Bare except clause	Warning	✅ Yes	✅ Yes
Mutable default argument	Warning	❌ No	✅ Yes
Wildcard import	Info	❌ No	✅ Yes
Security Issues
Dangerous `eval()` usage	Critical	❌ No	✅ Yes
Dangerous `exec()` usage	Critical	❌ No	✅ Yes
Unsafe pickle deserialization	High	❌ No	✅ Yes
Insecure module import	Info	❌ No	✅ Yes
Quality Issues
High cyclomatic complexity	Warning	❌ No	✅ Yes
Too many function arguments	Info	❌ No	✅ Yes
Missing docstrings	Info	❌ No	✅ Yes
Magic numbers	Info	❌ No	✅ Yes
Too many methods (God object)	Warning	❌ No	✅ Yes
Assert in production code	Info	❌ No	✅ Yes

Legend:

✅ Yes = Fully automated fix available
❌ No = Requires manual intervention or AI
AI-Fix = Can be fixed using AI (Claude/GPT)

🎯 Roadmap & Future Enhancements

Completed ✅

Automated fix application
AI-powered bug detection and fixing
Code quality metrics and complexity analysis
Comprehensive test suite
Configuration file support
Pull request automation

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
examples		examples
fixer		fixer
pr-handler		pr-handler
scanner		scanner
tests		tests
.gitignore		.gitignore
README.md		README.md
config.example.yaml		config.example.yaml
config.py		config.py
main.py		main.py
requirements.txt		requirements.txt
setup.py		setup.py

Folders and files

Latest commit

History

Repository files navigation

Bug-Fixing-AI-Assistant 🤖🔧

Project Structure

✨ Features

1. 🔍 Intelligent Code Scanning

2. 🤖 AI-Powered Fix Generation

3. ⚡ Automated Fix Application

4. ✅ Validation & Testing

5. 🔄 Pull Request Integration

🚀 Quick Start

Installation

Command Line Usage

Basic Scanning (No AI)

AI-Powered Fixing

Using Configuration File

Python API Usage

Simple Scanning

AI-Powered Fixing

Create Pull Request

Running Tests

Examples

Development

Project Requirements

Contributing

Architecture

Scanner Module

Fixer Module

PR Handler Module

📋 Supported Issue Types

🎯 Roadmap & Future Enhancements

Completed ✅

Planned 🚧

License

Contact

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages