Alice Releases Caterpillar After Catching Malicious OpenClaw Skills Used by 6,000+ Users

Free open-source scanner helps security teams and builders identify malicious behaviors as OpenClaw adoption accelerates

NEW YORK, Feb. 4, 2026 /PRNewswire/ — Alice today announced the release of Caterpillar, a free, open-source security tool built to help developers, security teams and users protect AI agents in OpenClaw (formerly ClawdBot and Moltbot) as agent ecosystems accelerate across the developer community. OpenClaw’s momentum reflects a broader industry shift: agents are quickly moving from experimentation to operational software and connecting to tools, calling APIs, and executing workflows with increasing autonomy.

Caterpillar is designed for the security reality that comes with that shift. It statically inspects skill logic and configurations to surface injection paths, unsafe tool access, and obfuscated behaviors. The software is further informed by RabbitHole, Alice’s adversarial intelligence database built from years of real-world threat research and signals.

Caterpillar’s release follows an early real world test case where the tool flagged several published skills Alice found to be actively malicious, including skills in use by over 6,000 OpenClaw users when they were caught.

“Agent ecosystems are scaling faster than the security assumptions around them,” said Noam Schwartz, CEO of Alice. “When you install a skill, you’re not installing a feature, you’re installing behavior. Caterpillar helps teams see what they’re actually running, and catch issues early, before they become incidents. For us, the goal is simple: help builders and businesses advance unafraid.”

Caterpillar is available today as free, open-source software, enabling the community to audit, extend, and operationalize agent security as adoption grows. For more information or to get started, visit caterpillar.alice.io.

Watch the video on Caterpillar here.

About Alice

Alice is a trust, safety, and security company built for the GenAI era. For nearly a decade Alice has safeguarded communicative technologies used by 3 billion people to create, collaborate, and interact – whether with each other or with machines. Alice transforms real-world abuse signals into intelligence that secures today’s leading enterprises and frontier model labs. Powered by Rabbit Hole, Alice delivers end-to-end safety and security across the AI lifecycle, from model hardening and red-teaming to runtime guardrails and ongoing drift detection. Today, Alice supports UGC platforms, frontier model labs, and enterprises building AI applications and agents worldwide. Our role is to provide the trusted safety and security layer that lets the companies shaping the future of AI advance unafraid.

For more information, please visit www.alice.io.

Media Contact:
Eden Hoffman
ehoffman@bambergervlasto.com

View original content to download multimedia:https://www.prnewswire.com/news-releases/alice-releases-caterpillar-after-catching-malicious-openclaw-skills-used-by-6-000-users-302679381.html

SOURCE Alice