open-education-hub-deep-dives

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

• Third-party content exposure detected (high risk: 0.80). The skill's required "Research Process" (see "External Research" under Research Process / references/proprietary-sources.md and SKILL.md) explicitly instructs the writer/agent to gather and integrate material from open/public third‑party sources (NHERI, NCES, HSLDA, influencer blogs/podcasts and other influencer content), which the agent is expected to read and use to shape article content and actions — creating a clear pathway for untrusted, user-generated web content to influence behavior.