ruvector-agentic-synth

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

• Third-party content exposure detected (high risk: 0.70). The skill’s runtime path gen.generateQA(documents, ...) ingests the caller-supplied documents: string[] (e.g., scraped/public web text or other users’ authored content) into the LLM to generate Q&A, which is outsider-authored free text when the operating user didn’t author/provide it.