research-project-orienter

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

• Third-party content exposure detected (high risk: 0.75). This skill reads outsider-authored free text from repository manifest files under .research/ (e.g., .research/decisions.md, .research/open_questions.md, and other YAML/MD content) and injects it into the LLM as the “orientation memo” context.