ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers

2026-03-25

AI Summary

This work addresses the security vulnerabilities of OpenClaw agents, which, due to their high privileges, are susceptible to sensitive data leakage, privilege escalation, and malicious skill execution. Existing defenses are fragmented and lack comprehensive coverage across the agent’s lifecycle. To mitigate these risks, we propose ClawKeeper—a real-time security framework grounded in a decoupled Watcher paradigm. ClawKeeper establishes a three-tiered defense architecture spanning instruction, runtime, and system layers through skill-level policy injection, pluggable runtime monitoring, and system-wide state validation. Crucially, it enables dynamic intervention without modifying the agent’s core logic. Our evaluation demonstrates that ClawKeeper significantly enhances security across diverse threat scenarios, and we release all code publicly to foster further research and adoption.

Technology Category

• Multiagent Systems: Adversarial Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Application Domains: Security

Application Category

• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search

Abstract

OpenClaw has rapidly established itself as a leading open-source autonomous agent runtime, offering powerful capabilities including tool integration, local file access, and shell command execution. However, these broad operational privileges introduce critical security vulnerabilities, transforming model errors into tangible system-level threats such as sensitive data leakage, privilege escalation, and malicious third-party skill execution. Existing security measures for the OpenClaw ecosystem remain highly fragmented, addressing only isolated stages of the agent lifecycle rather than providing holistic protection. To bridge this gap, we present ClawKeeper, a real-time security framework that integrates multi-dimensional protection mechanisms across three complementary architectural layers. (1) \textbf{Skill-based protection} operates at the instruction level, injecting structured security policies directly into the agent context to enforce environment-specific constraints and cross-platform boundaries. (2) \textbf{Plugin-based protection} serves as an internal runtime enforcer, providing configuration hardening, proactive threat detection, and continuous behavioral monitoring throughout the execution pipeline. (3) \textbf{Watcher-based protection} introduces a novel, decoupled system-level security middleware that continuously verifies agent state evolution. It enables real-time execution intervention without coupling to the agent's internal logic, supporting operations such as halting high-risk actions or enforcing human confirmation. We argue that this Watcher paradigm holds strong potential to serve as a foundational building block for securing next-generation autonomous agent systems. Extensive qualitative and quantitative evaluations demonstrate the effectiveness and robustness of ClawKeeper across diverse threat scenarios. We release our code.

Problem

• autonomous agents
• security vulnerabilities
• OpenClaw
• system-level threats
• holistic protection

Innovation

• Skill-based protection
• Plugin-based protection
• Watcher-based protection
• autonomous agent security
• real-time intervention