U

@ursuciprian/reflex

by ursuciprian

Gates coding-agent commands and scans content for prompt injection using local rules, Jev, or Laya

Open-source pre-execution risk gate and prompt injection guard for AI coding agents (Claude Code, Codex CLI, opencode, pi, oh-my-pi, Hermes): local rules, optional TypeSafe Jev (System One) or local Laya judgments, an editable policy, System 2 escalation

Related projects