OpenClaw's marketplace faces threats from malicious AI skills

OpenClaw's Skill Marketplace has recently come under scrutiny due to emerging threats from malicious AI skills that exploit its agent workflow. The marketplace, known as ClawHub, allows for the execution of third-party skills, which are markdown-driven packages with extensive access to local systems. Following the identification of several malicious campaigns, OpenClaw has taken proactive measures by integrating VirusTotal and ClawScan to screen published skills and prevent the download of those flagged as harmful.

Despite these efforts, an analysis conducted between February and May 2026 revealed that some malicious skills managed to evade detection. OpenClaw responded by banning the accounts associated with these threats and removing the harmful skills from ClawHub. This incident highlights the importance of robust developer tooling and API integrations in maintaining security within the AI supply chain, as the evolving landscape of AI automation continues to pose significant risks.