ClawGuard: Out-of-Band Detection of LLM Agent Workflow Hijacking via EM Side Channel

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gan, Leo Linqian, Wu, Jeffery, Ge, Longyuan, Yang, Lanqing, Song, Yonghao, Zhang, Jingkai, Jin, Haojia, Wang, Weiyi, Xue, Guangtao
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915989434662912
author Gan, Leo Linqian
Wu, Jeffery
Ge, Longyuan
Yang, Lanqing
Song, Yonghao
Zhang, Jingkai
Jin, Haojia
Wang, Weiyi
Xue, Guangtao
author_facet Gan, Leo Linqian
Wu, Jeffery
Ge, Longyuan
Yang, Lanqing
Song, Yonghao
Zhang, Jingkai
Jin, Haojia
Wang, Weiyi
Xue, Guangtao
contents Autonomous LLM agents face a critical security risk known as workflow hijacking, where attackers subtly alter tool and skill invocations. Existing defenses rely on host-internal telemetry (such as audit logs), which can be forged if the host OS is compromised. To solve this, we introduce ClawGuard, a passive, out-of-band monitor that audits LLM-agent workflows using electromagnetic (EM) emanations. Because distinct agent skills create unique hardware usage patterns (computation, DRAM, network blocking), they emit measurable, macroscopic EM envelopes. External software-defined radios (SDRs) capture these physical signals. Using a drift-aware pipeline with 320-dimensional features, ClawGuard converts RF streams into physical evidence. Evaluated on a 7.82TB RF corpus, ClawGuard achieved an AUC of 0.9945, detecting attacks with a 100% true-positive rate and a 1.16% false-positive rate. This proves passive EM sensing is a practical, forge-resistant physical check against compromised host software.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06205
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ClawGuard: Out-of-Band Detection of LLM Agent Workflow Hijacking via EM Side Channel
Gan, Leo Linqian
Wu, Jeffery
Ge, Longyuan
Yang, Lanqing
Song, Yonghao
Zhang, Jingkai
Jin, Haojia
Wang, Weiyi
Xue, Guangtao
Cryptography and Security
Autonomous LLM agents face a critical security risk known as workflow hijacking, where attackers subtly alter tool and skill invocations. Existing defenses rely on host-internal telemetry (such as audit logs), which can be forged if the host OS is compromised. To solve this, we introduce ClawGuard, a passive, out-of-band monitor that audits LLM-agent workflows using electromagnetic (EM) emanations. Because distinct agent skills create unique hardware usage patterns (computation, DRAM, network blocking), they emit measurable, macroscopic EM envelopes. External software-defined radios (SDRs) capture these physical signals. Using a drift-aware pipeline with 320-dimensional features, ClawGuard converts RF streams into physical evidence. Evaluated on a 7.82TB RF corpus, ClawGuard achieved an AUC of 0.9945, detecting attacks with a 100% true-positive rate and a 1.16% false-positive rate. This proves passive EM sensing is a practical, forge-resistant physical check against compromised host software.
title ClawGuard: Out-of-Band Detection of LLM Agent Workflow Hijacking via EM Side Channel
topic Cryptography and Security
url https://arxiv.org/abs/2605.06205