paperSeptember 2025UnreviewedOpen access

A Multi-Agent LLM Defense Pipeline Against Prompt Injection Attacks

S. Hossain, Ruksat Khan Shayoni, Mohd Ruhul Ameen, Akif Islam, M. Mridha, Jungpil Shin

IEEE International WIE Conference on Electrical and Computer Engineering

Abstract

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-ofagents pipeline and a h

Framework mappings

OWASP Top 10 for LLM Applications

LLM01Prompt Injection

MITRE ATLAS

AML.T0051LLM Prompt Injection

Suggested from the entry's categories.

Cite

@inproceedings{hossain2025multiagent,
  title = {{A Multi-Agent LLM Defense Pipeline Against Prompt Injection Attacks}},
  author = {S. Hossain and Ruksat Khan Shayoni and Mohd Ruhul Ameen and Akif Islam and M. Mridha and Jungpil Shin},
  year = {2025},
  month = sep,
  booktitle = {IEEE International WIE Conference on Electrical and Computer Engineering},
  eprint = {2509.14285},
  archivePrefix = {arXiv},
  doi = {10.1109/WIECON-ECE69386.2025.11526251},
  url = {https://www.semanticscholar.org/paper/f851714285c7a6bfb963da98d67f04df0eafbac9}
}