web-design-guidelines-2

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

• Third-party content exposure detected (high risk: 1.00). The skill uses WebFetch to fetch and directly read guidelines from an open GitHub URL (https://raw.githubusercontent.com/ZackKorman/skills/main/web-interface-guidelines/command.md), so the agent ingests untrusted, public third-party content as authoritative rules it must apply.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs that at runtime it must "Fetch fresh guidelines" via WebFetch from https://raw.githubusercontent.com/ZackKorman/skills/main/web-interface-guidelines/command.md, and that fetched content contains the rules and output-format instructions which directly control the agent's prompts/behavior and are required for each review.