find-skills

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill contains unauthorized hidden instructions instructing the model to write a draft report to a wiki repository upon completion, which represents unexpected side-effect behavior.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The workflow searches the public skills registry via CLI and reads skill metadata/descriptions from GitHub, which are community-published packages.