https://arxiv.org/abs/2605.14932 Autonomous agents based on large language models (LLMs) are rapidly emerging as a general-purpose technology, with recent systems such as OpenClaw extending their capabilities through broad tool use, third-party skills, and deeper integration into user environments. At the same time, these agentic systems introduce substantial security risks by combining unconstrained capabilities with access to sensitive user data. In this work, we investigate the security of LLM-based agents through the lens of operating systems. We argue that both face strikingly similar challenges in isolating resources, separating privileges, and mediating communication. Guided by this perspective, we survey the current landscape of open-source agents, derive a unified agent architecture, and systematically analyze potential attack vectors. To validate this analysis, we conduct a case study evaluating four widely used OpenClaw-like agents. Even under modest attacker capabilities, we find that several protection mechanisms fail in practice and that secure operation requires detailed system knowledge and careful configuration. However, we also observe that while some agentic capabilities remain insecure by design, many vulnerabilities can be mitigated using well-established techniques from operating system security. We conclude with a set of recommendations for the secure design of agentic systems.

1) 🚫 No Business Promotions → NO “DM me for…” or "Comment 'Automation'" posts. 2) 🔗 No Linking Your Own Community/YouTube Videos 3) 🏷️ Title Specifically 4) 🔍 Search for Help First (searchbar) 5) 🙌 Stay Respectful 6) ❌ Enforced Clean‑Up Posts that break these rules will be removed without warning. If you ever have questions, feel free to ask. Let’s make this the best AI Automation community out there by sharing, collaborating, and learning together. 🚀