michaelansel / CoreContributorBot

Use an LLM to maintain a code repository, acting as a core contributor to the codebase
0 stars 0 forks source link

GitHub Bot

This repository contains a GitHub bot that automatically processes issues and pull requests using an AI-powered RAG loop. The bot is implemented in Python and runs in a Docker container.

Features

Requirements

Setup

  1. Clone the repository
  2. Create a .env file based on the .env.sample file and add your GitHub and OpenAI API tokens
  3. Build the Docker image by running ./scripts/build.sh
  4. Run the unit tests by running ./scripts/test.sh
  5. Run the bot by running ./scripts/run.sh

Automated Testing

The repository includes a GitHub Actions workflow that automatically runs the build and test scripts on every pull request. The workflow is defined in the .github/workflows/test.yml file.

Pull requests can only be merged if the automated tests pass successfully.