Code Engine for Bug Fixing

A Python-based agent that helps non-technical users fix bugs and make changes to codebases through natural language requests.

This POC was created as part of the Dust AI Agents Hackathon in Paris.

Features

• Uses LLM to analyze and fix code issues
• Git integration for pulling, committing, and pushing changes
• File management capabilities (read/write)
• REST API for integration with other applications

Setup

1. Clone the repository
2. Install dependencies:

   pip install -r requirements.txt
   

3. Create a .env file with the following variables:

   MODEL_ID=your_model_name
   API_KEY=your_api_key
   CODEBASE_PATH=/path/to/codebase
   

Usage

As a command-line tool:

python agent.py "fix the button click handler in main.js"

As an API server:

python api.py

Then send POST requests to /api/run with JSON body:

{
  "query": "fix the button click handler in main.js",
  "context": "additional context about the application"
}

Technical Details

• Built with smolagents from HuggingFace
• Uses LiteLLM for model access
• Git operations handled through subprocess, relying on a user being authenticated on the system

Next steps

In order to improve the agent's capabilities for medium and large code base, we would need to have some kind of codebase embedding and RAG.