Newsletter Summary Writer

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly tells the agent to "Read the full source content" from arbitrary sources (e.g., "a long article, a Twitter/X thread, a blog post" or a URL), so the agent ingests untrusted public third‑party content that directly drives its summarization and could carry indirect prompt injection.