latent_affect

I developed a tool that can catch AI cheating. Is it viable for research? Like many people, I’m developing a harness for agentic AI to combat context bloat and enable autonomous code execution. It consists of three layers: - the core harness with hooks, scripts, and skills - a zero...