The Agent Skills Directory

[Unverifiable Dependencies & Remote Code Execution] (CRITICAL): Automated scanners detected a blacklisted malicious URL within the skill source (main.rs). The use of known-hostile domains for remote resource fetching is a definitive indicator of an attempt to execute malicious payloads or scripts.
[Data Exposure & Exfiltration] (CRITICAL): Connection attempts to blacklisted domains are frequently associated with the exfiltration of sensitive information, such as environment variables, system configuration, or user credentials, to attacker-controlled command-and-control infrastructure.

codebase-docs