audit-library-health

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly runs npx ai-agent-skills check --format json and the "Gotchas" state that check makes network requests to verify upstream sources, meaning the agent fetches and interprets potentially untrusted public repository/content which can influence subsequent health-check actions.