jev-guard

A Claude Code plugin that scores what the model is about to do — and what it just claimed to have done — with a probability model, denying the action outright when the risk score is high enough, moving rules that would otherwise be ignored once a CLAUDE.md gets buried by a long context into enforced hooks instead.

unverified — sourced from awesomejev.com; description independently written by AionEdge from the repo’s own GitHub metadata and README, not GitHub-API relevance-checked