cs.CRNov 18, 2026

AutoPI: Taxonomy-Guided Adaptive Red Teaming for Prompt Injection in LLM Agents

Authors: Stavros Damianakis, Konstantinos Psounis

Organizations: University of Southern California

Abstract

Prompt injection is the dominant unsolved security risk for LLM agents, and the community now agrees that robustness must be measured against adaptive attackers rather than fixed benchmark injections. Many adaptive red teamers exist, and most are built to maximize attack success rate (ASR). Yet ASR counts how often a target breaks, not in how many different ways. A red teamer can achieve a high ASR by reusing a few techniques that work, leaving the others untested and absent from the attacks used to harden models. We present AutoPI, an adaptive red teamer designed to find many different ways to break a target, not just to break it often. AutoPI uses PIAtlas, a taxonomy of prompt-injection techniques, both to guide its search and to measure coverage. For each scenario, an attacker LLM writes an injection for every sub-technique in the taxonomy and then reworks the ones that fail using feedback from an LLM judge. The method is black-box and training-free, it plugs into existing benchmarks in place of their fixed attack templates, and it stays up to date as the taxonomy evolves. We evaluate AutoPI on AgentDojo and IPI Arena against open-weight and proprietary target models. Alongside ASR, we report technique coverage, the fraction of taxonomy classes, techniques, and sub-techniques that yield at least one successful injection. On AgentDojo, AutoPI finds a successful injection against Qwen3.6-35B in all six classes of the taxonomy and for more than 90% of its techniques. Against Claude Haiku 5.5, the most robust target we evaluate, it achieves between 20% and 50% higher ASR and between 25% and 400% higher class-level coverage than recent state-of-the-art adaptive red teamers, while requiring no training. We release our code and the generated attacks to support reproducible, coverage-aware evaluation.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. PI-Hunter: Automated Red-Teaming for Exposing and Localizing Prompt Injections

    Jun 10, 2026Pengfei He, Lesly Miculicich, Vishesh Sharma +5AI Agent AuditingLLM Auditing

  2. PISmith: Reinforcement Learning-based Red Teaming for Prompt Injection Defenses

    Mar 13, 2026Chenlong Yin, Runpeng Geng, Yanting Wang +1Prompt Injection Attacks on LLMsLLM Red Teaming

  3. PIPETTE: Prompt Injection Purple-Teaming for Robust LLM Agents

    Nov 18, 2026Lipeng He, Yihan Wang, Jiawen Zhang +1Prompt Injection Attacks on AI AgentsPrompt Injection Defense