gj0xv

Uses Clef to decide whether a prompt needs a frontier model or a cheap one, behind an OpenAI-compatible proxy and library to cut spend.

Has Clef score retrieved chunks against the question and keeps only the evidence, shrinking noisy RAG context before it reaches the LLM.

Routes support tickets by intent, scores urgency, frustration and churn risk, and escalates by confidence, all with a local System 1 model.

Toolkit for judging outputs cheaply with Clef and auditing whether its confidence numbers are calibrated, with a rendered demo clip.