GitHub and its parent company Microsoft are now part of the steering committee for MCP (Model Connection Protocol), a standard set by Anthropic aimed at linking AI models with data systems. This announcement was made during Microsoft’s Build 2025 conference, signalling a growing momentum for MCP within the AI sector. Earlier in the year, major players like OpenAI and Google also pledged their support for MCP across their AI offerings.
MCP is designed to facilitate data retrieval from various sources, including business applications and software, allowing AI models to perform tasks efficiently. It enables the creation of two-way connections between data sources and AI applications, such as chatbots. Developers can set up “MCP servers” to expose their data and create “MCP clients” — applications or workflows that connect to these servers as needed.
Microsoft and GitHub plan to provide extensive first-party support for MCP through their ecosystems, which includes services like Microsoft Azure and platforms like Windows 11. Notably, Windows is set to introduce MCP integrations in the coming months. These integrations will allow developers to expose app functionalities to models that are MCP-enabled. Microsoft outlined that developers could wrap key features of their applications as MCP servers, allowing interaction with Windows system functionalities such as the file system and the Windows Subsystem for Linux.
Moreover, Microsoft and GitHub are contributing to the development of the MCP standard. Microsoft’s identity and security teams have collaborated with Anthropic and the wider MCP community to create a refined authorization specification, enabling more secure connections between MCP-connected applications and servers through trusted login methods. This aims to enhance access control for AI applications dealing with data from personal storage and subscription services.
Additionally, GitHub has partnered with the MCP steering committee to develop a registry service for MCP servers. This service will enable developers to manage public or private centralized repositories for MCP server entries, simplifying the discovery and management of various MCP implementations and their configurations.
In summary, the collaboration between GitHub, Microsoft, and the MCP steering committee marks a significant step toward standardising connections between AI models and data sources, enhancing the functionality and security of AI applications in diverse business environments.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence
