spec-kitty-tasks

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

• Third-party content exposure detected (high risk: 0.90). The skill's research task workflow (templates/tasks.md and templates/tasks-template.md) explicitly instructs agents to search and ingest public third‑party sources (e.g., "Search academic database 1 (IEEE, PubMed, arXiv, etc.)" and "Search gray literature and industry sources") and to extract findings that drive subtasks and decisions, so untrusted external content is read and can materially influence agent actions.