gathering-competitive-intel

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

• Third-party content exposure detected (high risk: 1.00). This skill's workflow explicitly instructs the AI to crawl and extract from public sources such as competitor websites, G2/Capterra reviews, Reddit/community discussions, Google search results, Crunchbase/news, Twitter/LinkedIn, and product changelogs (see "Data Sources" and "Data Collection Methods" in workflow.md), meaning the agent ingests untrusted third-party/user-generated content that can trigger automated alerts and update actions.