unfairgaps

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

• Third-party content exposure detected (high risk: 0.90). The skill's native flow explicitly runs WebSearch + WebFetch (SKILL.md "Native flow" and Phase 3 Evidence ledger) to fetch open/public web pages and PDFs (regulator sites, court dockets, news/aggregator/blog sources listed in the query and source_class rules) and then reads and ingests those pages into evidence cards that directly drive decisions (unfairgap status, product sketches, pitch templates), so it clearly consumes untrusted third‑party web content as part of its required workflow.