twitter-article-reader

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to construct jina.ai proxy URLs (https://r.jina.ai/https://x.com/...) and use web_fetch to retrieve and return Twitter/X user-generated content, so the agent will ingest and interpret untrusted third‑party social‑media content that could contain malicious instructions.