unknown-patterns

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests external package documentation via "erudita install" / "erudita fetch --deps" into the .erudita directory and then runs semantic searches with "osgrep search", meaning the agent will read and interpret third-party (untrusted) docs from external dependencies.