exploring-app-ideas

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow and scripts (see SKILL.md "Competitor Gap Analysis" / "Research Sources", scripts/analyze-competitors.js research guide, and validate-market.js) explicitly instruct researching and ingesting public, user-generated sources such as Reddit, Product Hunt, Twitter, App Store/G2/Capterra reviews and other forums — third-party content that the agent would read and that can materially influence analysis and next actions, opening the possibility of indirect prompt injection.