babeliocli-library

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md shows the skill uses babeliocli to fetch the authenticated user's Babelio pages/shelves (public, user-generated content on babelio.com) and the agent consumes fields like titles, tags, URLs and dates as part of its workflows (pagination and book listing are explicit), so untrusted third‑party content can influence filtering, selection, and subsequent actions.