add-golden

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts arbitrary URLs and in Phase 1-2 ("Input Collection" / "Fetch and Extract") fetches and parses content from web pages (examples: /add-golden https://example.com/article, https://arxiv.org/...), so it ingests untrusted public third-party content that the agents read and analyze.