Lune

EMNLP Findings2025

DemonAgent: Dynamically Encrypted Multi-Backdoor Implantation Attack on LLM-based Agent

Pengyu Zhu, Zhenhong Zhou, Yuanhe Zhang, Shilinlu Yan, Kun Wang, Sen Su

Abstract

As LLM-based agents become increasingly prevalent, triggers implanted in user queries or environment feedback can activate hidden backdoors, raising critical concerns about safety vulnerabilities in agents. However, traditional backdoor attacks are often detectable by safety audits that analyze the reasoning process of agents, hindering further progress in agent safety research. To this end, we propose a novel backdoor implantation strategy called Dynamically Encrypted Multi-Backdoor Implantation Attack. Specifically, we introduce dynamic encryption, which maps the backdoor into benign content, effectively circumventing safety audits. To enhance stealthiness, we further decompose the backdoor into multiple subbackdoor fragments. Based on these advancements, backdoors are allowed to bypass safety audits significantly. Additionally, we present AgentBackdoorEval, a dataset designed for the comprehensive evaluation of agent backdoor attacks. Experimental results across multiple datasets demonstrate that our method achieves an attack success rate approaching 100% while maintaining a detection rate of 0%, illustrating its effectiveness in evading safety audits. Our findings highlight the limitations of existing safety mechanisms in detecting advanced attacks, underscoring the urgent need for more robust defenses against backdoor threats. Code and data are available at https: //github.com/whfeLingYu/DemonAgent.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

2 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs cites Toolformer: Language Models Can Teach Themselves to Use Tools
  • ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs cites HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
  • HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face cites Toolformer: Language Models Can Teach Themselves to Use Tools
  • AdapAction: Adaptive Target Action Backdoor Attack against GUI Agents cites WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers2

  1. When "Correct" Is Not Safe: Can We Trust Functionally Correct Patches Generated by Code Agents?Yibo Peng, James Song, Lei Li, Xinyu Yang et al.ACL 2026 · 2 citations
  2. AdapAction: Adaptive Target Action Backdoor Attack against GUI AgentsBaicheng Chen, Mingda Zhang, Min Zhang, Haizhou Li et al.CVPR 2026

Builds on15

  1. Toolformer: Language Models Can Teach Themselves to Use ToolsTimo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu et al.NeurIPS 2023 · 5,989 citations
  2. Generative Agents: Interactive Simulacra of Human BehaviorJoon Sung Park, Joseph C. O'Brien, Carrie Jun Cai, Meredith Ringel Morris et al.UIST 2023 · 1,882 citations
  3. HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging FaceYongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li et al.NeurIPS 2023 · 1,778 citations
  4. WebShop: Towards Scalable Real-World Web Interaction with Grounded Language AgentsShunyu Yao, Howard Chen, John Yang, Karthik NarasimhanNeurIPS 2022 · 1,477 citations
  5. ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIsYujia Qin, Shihao Liang, Yining Ye, Kunlun Zhu et al.ICLR 2024 · 1,469 citations

Related papers

  1. Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based AgentsWenkai Yang, Xiaohan Bi, Yankai Lin, Sishuo Chen et al.NeurIPS 2024 · 195 citations
  2. BadAgent: Inserting and Activating Backdoor Attacks in LLM AgentsYifei Wang, Dizhan Xue, Shengjie Zhang, Shengsheng QianACL 2024
  3. AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge BasesZhaorun Chen, Zhen Xiang, Chaowei Xiao, Dawn Song et al.NeurIPS 2024 · 539 citations
  4. BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based AgentsYunhao Feng, Yige Li, Yutao Wu, Yingshui Tan et al.ACL Findings 2026 · 1 citation
  5. Hidden Ghost Hand: Unveiling Backdoor Vulnerabilities in MLLM-Powered Mobile GUI AgentsPengzhou Cheng, Haowen Hu, Zheng Wu, Zongru Wu et al.EMNLP Findings 2025 · 1 citation
Dusk over the sea between two cliffs drawn in fine vertical lines