find-twitter-influencers

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

• Third-party content exposure detected (high risk: 0.85). High: the workflow uses Exa to fetch arbitrary web pages at runtime (api:"exa" with contents.text), then parses that outsider-authored page text (listicles/articles by third parties) to extract Twitter handles and bios, which becomes LLM-readable context for subsequent steps.