{"skill":{"slug":"runtime-monitor","displayName":"Runtime Monitor","summary":"AI 代理运行时 I/O 安全监控 - 检测提示注入、数据外泄、危险命令","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":123,"installsAllTime":1,"installsCurrent":1,"stars":0,"versions":1},"createdAt":1774651983907,"updatedAt":1774652509181},"latestVersion":{"version":"1.0.0","createdAt":1774651983907,"changelog":"- Initial release of runtime-monitor skill.\n- Monitors AI agent runtime I/O for security threats.\n- Features detection for prompt injection, data leakage, and dangerous system commands.\n- Provides risk classification (LOW, MEDIUM, HIGH, CRITICAL) for easy triage.\n- Includes integration guide and best practices for optimal security and auditing.","license":"MIT-0"},"metadata":null,"owner":{"handle":"liwiw","userId":"s17anbsp7qj1ebpw38c4nkrmwd83fc3g","displayName":"liwiw","image":"https://avatars.githubusercontent.com/u/266823704?v=4"},"moderation":null}