I’m building OSSAIX (https://ossaix.com ) — a curated directory for open-source AI projects.

The goal is to make it easier for developers to discover and evaluate OSS AI tools (LLMs, RAG/agents, local AI, image/audio/video, etc.) without digging through endless GitHub repos.

Each project is reviewed and includes basic signals like categories and GitHub activity to help assess usefulness quickly.

This is still early, and I’d love feedback from the HN community:

Is a curated OSS AI directory useful?

What signals/features would help you decide faster?

What’s missing or unnecessary?

Thanks in advance for any thoughts