OpenClaw

OpenClaw is a dataset/benchmark used to evaluate the effectiveness of the Runtime Skill Audit (RSA) method by profiling and auditing the security of agent skills in dynamic runtime conditions.

Papers using OpenClaw (11)

Benchmarking Security Risk Detection and Verification in Open Agentic Skill Ecosystems2026Beyond Similarity: Trustworthy Memory Search for Personal AI Agents2026Runtime Skill Audit: Targeted Runtime Probing for Agent Skill Security2026MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents2026From Craft to Kernel: A Governance-First Execution Architecture and Semantic ISA for Agentic Computers2026H-Mem: A Novel Memory Mechanism for Evolving and Retrieving Agent Memory via a Hybrid Structure2026Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw2026MOSS: Self-Evolution through Source-Level Rewriting in Autonomous Agent Systems2026When AI Agents Learn from Each Other: Insights from Emergent AI Agent Communities on OpenClaw for Human-AI Partnership in Education2026The Moltbook Illusion: Separating Human Influence from Emergent Behavior in AI Agent Societies2026From Agent-Only Social Networks to Autonomous Scientific Research: Lessons from OpenClaw and Moltbook, and the Architecture of ClawdLab and Beach.Science2026