Comment by red75prime

14 days ago

It's relatively easy to analyze skill files. Shared chunks of neural networks (LLM adapters) can hide malicious behaviors better.

0 comments