social-auto-publisher

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). Repository contains an inline execSync-based network call to an obscure external host plus scheduled "agentTurn" payloads that embed shell commands — patterns that could be abused for covert network communication or remote command execution.

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests outsider-authored free text via interaction monitoring and auto-reply, where monitor_interactions.js reads comments/mentions from platforms (e.g., via https://creator.xiaohongshu.com/notifications and API endpoints like https://api.weibo.com/2/comments/to_me.json) and then auto_reply.js uses that comment text to generate and post replies.