[2508.00500] ProbGuard: Probabilistic Runtime Monitoring for LLM Agent Safety

Computer Science > Artificial Intelligence arXiv:2508.00500 (cs) [Submitted on 1 Aug 2025 (v1), last revised 27 Mar 2026 (this version, v3)] Title:ProbGuard: Probabilistic Runtime Monitoring for LLM Agent Safety Authors:Haoyu Wang, Christopher M. Poskitt, Jiali Wei, Jun Sun View a PDF of the paper titled ProbGuard: Probabilistic Runtime Monitoring for LLM Agent Safety, by Haoyu Wang and Christopher M. Poskitt and Jiali Wei and Jun Sun View PDF HTML (experimental) Abstract:Large Language Model (LLM) agents increasingly operate across domains such as robotics, virtual assistants, and web automation. However, their stochastic decision-making introduces safety risks that are difficult to anticipate during execution. Existing runtime monitoring frameworks, such as AgentSpec, primarily rely on reactive safety rules that detect violations only when unsafe behavior is imminent or has already occurred, limiting their ability to handle long-horizon dependencies. We present ProbGuard, a proactive runtime monitoring framework for LLM agents that anticipates safety violations through probabilistic risk prediction. ProbGuard abstracts agent executions into symbolic states and learns a Discrete-Time Markov Chain (DTMC) from execution traces to model behavioral dynamics. At runtime, the monitor estimates the probability that future executions will reach unsafe states and triggers interventions when this risk exceeds a user-defined threshold. To improve robustness, ProbGuard incorporates s...