{
  "id": 16968,
  "url": "https://arxiv.org/abs/2608.05108v1",
  "title": "Agent Against Agent: An Agentic System for Automatic Prompt Injection Red Teaming",
  "summary": "Prompt injection poses significant security risks to LLM agents. Efficient and effective red-teaming is therefore critical, both for evaluating these risks and for collecting training data to improve defenses. Existing state-of-the-art prompt injection red-teaming methods primarily rely on reinforcement learning (RL), producing attacker models that often generalize poorly to new target LLMs. In this work, we develop PIMiner, an agentic system for prompt injection red-teaming. During training, PI",
  "authors": "Yanting Wang, Chenlong Yin, Runpeng Geng, Jinyuan Jia",
  "category": "research",
  "topics": "safety-alignment,agents-autonomy",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-05T17:44:09.000Z",
  "fetched_at": "2026-08-06T05:10:11.148Z",
  "source_slug": "x-arxiv-red-teaming-query",
  "source_name": "arXiv red teaming query",
  "source_homepage": "https://arxiv.org/a/redteam",
  "ethics_ai_record_url": "https://ethics.ai/record/16968",
  "original_url": "https://arxiv.org/abs/2608.05108v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}