AI-Infra-Guard (A.I.G) is a platform developed by Tencent that focuses on identifying security vulnerabilities in AI infrastructure. It offers tools like AI infrastructure vulnerability scans, MCP Server & Agent Skills scans, and Jailbreak Evaluations. The latest version, A.I.G V3.6.2, now supports n8n and includes detection for 78 Common Vulnerabilities and Exposures (CVEs) across 15 AI components, such as Ollama. Users can quickly get started with the service through Docker, with methods for one-click installation or building from the source. The platform is designed for internal use, lacking authentication for public deployment. It provides a modern web interface for user-friendly interaction, complete API documentation, and support for multiple operating systems, including Linux, macOS, and Windows. A.I.G is open-source and free under the MIT license, encouraging community contributions through an extensible plugin framework. This allows users to add their own fingerprint rules, vulnerability scan rules, and datasets for Jailbreak evaluations. The platform also emphasizes community involvement, urging users to contribute code, report issues, and engage in discussions. There’s a Pro version available for contributors, which includes advanced features and improved performance. A.I.G has been used in various academic papers, highlighting its relevance in AI security research. The article includes extensive details on installation, features, and community engagement, making it a comprehensive resource for anyone interested in AI security.