Make any LLM talk like a normal person. A system prompt that removes AI slop.
Repository Index
103,113 visible repositories in the index
1–3 of 3 · Most stars
A clinical decision-evidence benchmark: pull the one fact an answer rests on, and see whether the assistant's action moves with the evidence. Built on HealthBench.
Caveman Compression is a semantic compression method for LLM contexts. It removes predictable grammar while preserving the unpredictable, factual content that defines meaning.