Guardians of the Agentic System: Preventing Many Shots Jailbreak with Agentic System

TL;DR

This paper develops a new evaluation framework and anti-jailbreaking system for large language model agents, addressing security threats like many-shot jailbreaking and deceptive alignment, with promising detection accuracy but persistent vulnerabilities under long attacks.

Contribution

It introduces a comprehensive security framework combining detection and countermeasures for LLM agents, emphasizing active monitoring and adaptable interventions to enhance robustness.

Findings

94% detection accuracy on GEMINI 1.5 pro

Persistent vulnerabilities increase with attack length

Active monitoring improves security resilience

Abstract

The autonomous AI agents using large language models can create undeniable values in all span of the society but they face security threats from adversaries that warrants immediate protective solutions because trust and safety issues arise. Considering the many-shot jailbreaking and deceptive alignment as some of the main advanced attacks, that cannot be mitigated by the static guardrails used during the supervised training, points out a crucial research priority for real world robustness. The combination of static guardrails in dynamic multi-agent system fails to defend against those attacks. We intend to enhance security for LLM-based agents through the development of new evaluation frameworks which identify and counter threats for safe operational deployment. Our work uses three examination methods to detect rogue agents through a Reverse Turing Test and analyze deceptive alignment through multi-agent simulations and develops an anti-jailbreaking system by testing it with GEMINI 1.5 pro and llama-3.3-70B, deepseek r1 models using tool-mediated adversarial scenarios. The detection capabilities are strong such as 94\% accuracy for GEMINI 1.5 pro yet the system suffers persistent vulnerabilities when under long attacks as prompt length increases attack success rates (ASR) and diversity metrics become ineffective in prediction while revealing multiple complex system faults. The findings demonstrate the necessity of adopting flexible security systems based on active monitoring that can be performed by the agents themselves together with adaptable interventions by system admin as the current models can create vulnerabilities that can lead to the unreliable and vulnerable system. So, in our work, we try to address such situations and propose a comprehensive framework to counteract the security issues.