On the Effectiveness of Kernel-Level Evidence for Agent Security
Quick summary
arXiv:2609.28915v1 Announce Type: cross Abstract: LLM agents are deployed into infrastructure that grants them broad host authority, yet existing agent-security benchmarks and defenses operate almost exclusively at the application telemetry layer: the served tool manifest, the user prompt, and the model's messages. Some threats, however, smuggle malicious instructions and actions past the application boundary, leaving them invisible to that layer. In this work, we bridge that gap by pairing application-level agent telemetry with kernel-level syscall traces to present the first paired-evidence
Key takeaways
- arXiv:2609.28915v1 Announce Type: cross Abstract: LLM agents are deployed into infrastructure that grants them broad host authority, yet existing agent-security benchmarks and defenses operate almost exclusively at the application telemetry layer: the served tool manifest, the user prompt, and the model's messages.
- Some threats, however, smuggle malicious instructions and actions past the application boundary, leaving them invisible to that layer.
- In this work, we bridge that gap by pairing application-level agent telemetry with kernel-level syscall traces to present the first paired-evidence
Why it matters
“On the Effectiveness of Kernel-Level Evidence for Agent Security” shows why AI risk cannot be reduced to answer accuracy. Access controls, logging, human approval and incident response need to be designed into the workflow from the start.
Member comments