influencer-discovery

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

• Third-party content exposure detected (high risk: 1.00). Yes — the SKILL.md workflow (Steps 3–6 and the cross-platform discovery steps) explicitly instructs searching and ingesting public third-party sources such as YouTube, X/Twitter, blogs/Substack, podcasts, Instagram/Facebook, and conference speaker lists, which the agent must read and use to score and select influencers, exposing it to untrusted user-generated content that could carry indirect prompt injection.