# Researchers identify 627 malicious ClawHub skills using instruction and code analysis

_Published Tuesday, October 27, 2026 at 12:12 AM EDT · Science, Security · Latest · Tier 2 — Notable_

Researchers report identifying 627 malicious skills among 66,157 ClawHub skills after screening with a new security analysis framework and confirming findings through auxiliary dynamic validation and manual review. Their preprint introduces the Supply-Chain Scanner for Agent Skills, which examines natural-language instructions alongside executable code.

The scanner translates instructions into a constrained Python-style representation, then links their dependencies with those of bundled code to trace security risks across both. On a manually verified benchmark containing 117 malicious and 117 benign skills, it achieved average recall of 95.2% and accuracy of 85.6% across five language models. Precision was 80.2%, and average token consumption was 13.7 million.

## Sources

- [cs.CR updates on arXiv.org](https://arxiv.org/abs/2610.33489)

---
Canonical: https://techandbusiness.org/newswire/_udMFcFxETzhQ7J0ttpEwM
Published: 2026-10-27T04:12:50.426Z
Story chronology: 2026-10-27T04:00:00.000Z
Retrieved: 2026-10-27T06:36:18.683Z
Publisher: Tech & Business (techandbusiness.org)