Samuel Judson

Summary: Trail of Bits researchers Samuel Judson and Tjaden Hess report that they bypassed ClawHub's malicious skill detector, Cisco's agent skill scanner, and all three scanners integrated into skills.sh. The post is filed under machine-learning, vulnerabilities, and supply-chain.
What's compelling: Three separate scanner products and a platform's full set of integrated detectors were all evaded by one research team, which means the tooling currently standing between your organization and a malicious agent skill has a demonstrated bypass.

Read the original at Trail of Bits Blog ↗