seojoonkim

600+ pattern AI agent security defense covering prompt injection, supply chain injection, memory poisoning, action gate bypass, unicode steganography, and cascade amplification. Optional API for early-access and premium patterns. Tiered loading, hash cache, 11 SHIELD categories, 10 languages.

Runtime security layer for autonomous AI agents. Covers 6 attack surfaces: prompt injection, supply chain injection, memory poisoning, action gate bypass, unicode steganography, and cascade amplification. Collective immunity network — when one agent detects an attack, all connected agents become immune. 600+ patterns, token splitting bypass defense with 6-pass normalization.