researchers-verifier

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime workflow ingests “Research compiled into RESEARCH.md and SOURCES.md” produced by “Specialized Researchers” (other agents), so any outsider-authored free text in those files could be fed into the verifier’s LLM context for checking (indirect prompt-injection risk via researcher-provided content).