audit-imports

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

• Third-party content exposure detected (high risk: 0.70). The skill's Phase 2 explicitly fetches and parses public package metadata and files from third-party sites (e.g., https://registry.npmjs.org//latest, https://unpkg.com/@/..., and raw.githubusercontent.com) and then uses those parsed exports to drive analysis and automated edits, so untrusted, user-published content can directly influence agent decisions and tool actions.