parallel-deep-research

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs creating Parallel AI "research" tasks that perform "multi-source synthesis with citations" and return Markdown reports (see "Comprehensive research reports" and the Create Research Task / Example sections), which indicates the service fetches and ingests public third-party sources that the agent will read and could influence subsequent decisions.