skill-discovery

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to fetch remote READMEs and "record installation commands verbatim" into its output, which forces inclusion of any secrets or API tokens present in those commands.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches the remote README for L-yifan/skills at runtime and uses its contents (including raw installation commands) to drive candidate selection and outputs, so external repo content can directly control agent behavior.