leanplan

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

• Third-party content exposure detected (medium risk: 0.30). Leanplan’s runtime workflow ingests and LLM-reads a plan from the local filesystem via auto-discovery/search paths (docs/plans/, docs/superpowers/specs/, or *plan*.md) before analysis, meaning outsider-authored plan text can be included if an outsider can submit it into those files.