A Multi-Agent LLM Defense Pipeline Against Prompt Injection Attacks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912770193096704 |
|---|---|
| author | Hossain, S M Asif Shayoni, Ruksat Khan Ameen, Mohd Ruhul Islam, Akif Mridha, M. F. Shin, Jungpil |
| author_facet | Hossain, S M Asif Shayoni, Ruksat Khan Ameen, Mohd Ruhul Islam, Akif Mridha, M. F. Shin, Jungpil |
| contents | Prompt injection attacks represent a major vulnerability in Large Language Model (LLM) deployments, where malicious instructions embedded in user inputs can override system prompts and induce unintended behaviors. This paper presents a novel multi-agent defense framework that employs specialized LLM agents in coordinated pipelines to detect and neutralize prompt injection attacks in real-time. We evaluate our approach using two distinct architectures: a sequential chain-of-agents pipeline and a hierarchical coordinator-based system. Our comprehensive evaluation on 55 unique prompt injection attacks, grouped into 8 categories and totaling 400 attack instances across two LLM platforms (ChatGLM and Llama2), demonstrates significant security improvements. Without defense mechanisms, baseline Attack Success Rates (ASR) reached 30% for ChatGLM and 20% for Llama2. Our multi-agent pipeline achieved 100% mitigation, reducing ASR to 0% across all tested scenarios. The framework demonstrates robustness across multiple attack categories including direct overrides, code execution attempts, data exfiltration, and obfuscation techniques, while maintaining system functionality for legitimate queries. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14285 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Multi-Agent LLM Defense Pipeline Against Prompt Injection Attacks Hossain, S M Asif Shayoni, Ruksat Khan Ameen, Mohd Ruhul Islam, Akif Mridha, M. F. Shin, Jungpil Cryptography and Security Machine Learning Prompt injection attacks represent a major vulnerability in Large Language Model (LLM) deployments, where malicious instructions embedded in user inputs can override system prompts and induce unintended behaviors. This paper presents a novel multi-agent defense framework that employs specialized LLM agents in coordinated pipelines to detect and neutralize prompt injection attacks in real-time. We evaluate our approach using two distinct architectures: a sequential chain-of-agents pipeline and a hierarchical coordinator-based system. Our comprehensive evaluation on 55 unique prompt injection attacks, grouped into 8 categories and totaling 400 attack instances across two LLM platforms (ChatGLM and Llama2), demonstrates significant security improvements. Without defense mechanisms, baseline Attack Success Rates (ASR) reached 30% for ChatGLM and 20% for Llama2. Our multi-agent pipeline achieved 100% mitigation, reducing ASR to 0% across all tested scenarios. The framework demonstrates robustness across multiple attack categories including direct overrides, code execution attempts, data exfiltration, and obfuscation techniques, while maintaining system functionality for legitimate queries. |
| title | A Multi-Agent LLM Defense Pipeline Against Prompt Injection Attacks |
| topic | Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2509.14285 |