The Dark Side of LLMs: Agent-based Attack Vectors for System-level Compromise
- 18 August 2026
- book chapter
- Published by Springer Nature
- p. 5-23
Abstract
No abstract availableKeywords
All Related Versions
This publication has 17 references indexed in Scilit:
- TRiSM for Agentic AI: A review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent SystemsAI Open, 2026
- CyberRAG: An agentic RAG cyber attack classification and reporting toolFuture Generation Computer Systems, 2025
- Conversational health agents: a personalized large language model-powered agent frameworkJAMIA Open, 2025
- AGENTVIGIL: Automatic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents2025
- BadAgent: Inserting and Activating Backdoor Attacks in LLM Agents2024
- Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection2024
- Instructions as Backdoors: Backdoor Vulnerabilities of Instruction Tuning for Large Language Models2024
- Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection2023
- Backdoor Attacks on Pre-trained Models by Layerwise Weight Poisoning2021
- Weight Poisoning Attacks on Pretrained Models2020