digital-product-inventor

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

• Third-party content exposure detected (high risk: 0.90). The skill's Phase 2 "Market Signal Scan" (references/market-signal-scan.md and SKILL.md Phase 2) explicitly requires searching and ingesting content from public user-generated sources (Reddit, Discord, Twitter/X, Gumroad, Fiverr, Upwork) and using those posts/threads as evidence that informs signal scores, pricing, and go/no-go product decisions, which exposes the agent to untrusted third-party content that could carry indirect prompt-injection.