academic-pipeline

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly requires the integrity_verification_agent to perform WebSearch/DOI/ Semantic Scholar API lookups and to verify references, citation contexts, and claims by fetching publisher pages and public sources (see agents/integrity_verification_agent.md and SKILL.md Stage 2.5/4.5), and those external results are used as blocking verdicts that drive pipeline decisions—exposing the agent to untrusted third‑party content that could influence tool use.