StealthyPoison: Manipulating LLM Agent Behavior through Procedural Poisoning of Third-Party Skills

arXiv:2611.12155 · cs.CR · Submitted 2026-11-09 · Read on arXiv

cs.CR

Submitted: 2026-11-09

Updated: 2026-11-09

Code: https://github.com/cisco-ai-defense/skill-scanner

Terminology

Sources

Related papers