spark-recipe-newsletter-cleanup

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly reads email threads from the user's Inbox (commands: "spark emails Inbox" and "spark thread " in SKILL.md) and may follow mailing-list unsubscribe mechanisms, so it ingests untrusted third-party (external/user-generated) email content that can materially influence actions like unsubscribe, block, or enable autosummary.