ingesting-academic-content

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests arbitrary web pages (see SKILL.md "Web" ingestion and scripts/ingest-web.js) and the agent parses and uses that untrusted third‑party content (readability extraction → classification/structure → downstream parsing), so content from arbitrary URLs could carry instructions that materially influence processing and actions.