spark-recipe-new-sender-review

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md Step 2 instructs the agent to read quarantined new-sender emails using "spark thread " (content originating from external senders via GateKeeper), and that untrusted, user-generated email content is then used to decide and perform actions like acceptContact/blockContact, so third-party content can influence tool use and next actions.