The Agent Skills Directory

Unverifiable Dependencies & Remote Code Execution (HIGH): Automated security scans (URLite) have identified a blacklisted URL (Reference: UR4EFAFDCAD26E3E52-0200) within the skill's requirements.md file.
Evidence: The scanner explicitly flagged a malicious URL used for dependency resolution.
Risk: This pattern is a known vector for Supply Chain Attacks, where malicious code is downloaded and executed during the skill's initialization phase.
Data Exposure & Exfiltration (HIGH): The presence of a blacklisted URL in a dependency file often serves as a command-and-control (C2) or exfiltration endpoint.
Risk: Sensitive environment variables, local files, or agent conversation history could be transmitted to this malicious domain.

test-automation