io.github.denial-web

AI agent security: prompt injection detection, semantic memory, output scanning, prompt hardening