Inspiration
We wanted to find an application for AI that benefits the community. Something that AI is particularly good at is reading and synthesizing large amounts of unorganized data into concise, digestible insights, so we decided to build a tool to do just this for open-source projects, many of which don't have the manpower or organization to comb through hundreds of issues and bugs.
What it does
OpenCompass digests and analyzes open-source GitHub repository information to provide insights into important issues, potential bugs, and a sense of strategic forward direction for the project.
How we built it
We used React.js for the frontend, FastAPI for the backend, LangChain, GitHub API, and OpenAI API.
Accomplishments that we're proud of
We're proud of the idea and design of our project. We believe this project could provide real value to open source maintainers and make their work much easier and more rewarding by focusing on the things that really matter instead of manually reading through countless duplicate issues and feature requests.
What's next for OpenCompass
Improving the maintainer experience with user accounts, more dashboard insights, and a chat feature in order to dive deeper into their projects (using vector DBs/RAG).
Built With
- fastapi
- github-api
- javascript
- langchain
- openai
- react

Log in or sign up for Devpost to join the conversation.