Docker Launches MCP Catalog for Containerized AI Resources

Docker has announced the upcoming launch of the MCP Catalog, a new online directory aimed at facilitating the exchange of artificial intelligence (AI) resources. Set to debut in May, this platform will serve as a repository for AI servers and clients that can connect via the Model Context Protocol (MCP).

The MCP Catalog is designed to operate similarly to Docker Hub, which underpins its infrastructure. Users will be able to integrate MCP tools packaged in containers into their applications with just a few clicks, or alternatively, offer their own servers. The platform also automates processes for verification, authentication, and security.

For developers, the MCP Catalog will act as a centralized and reliable resource hub for discovering and sharing AI tools. Upon launch, Docker plans to feature over a hundred verified MCP servers from notable companies such as Stripe, Elastic, Heroku, Pulumi, Grafana Labs, Kong, Neo4j, New Relic, and Continue.

Tools encapsulated in containers can be quickly incorporated into applications to enhance specific AI functionalities. At launch, clients will be available for various platforms, including Docker AI Agent, Claude, Cursor, VS Code, Windsurf, Continue, and Goose, all with easy installation as depicted in promotional materials.

The catalog utilizes the same framework as Docker Hub, allowing organizations to regulate access to resources for their employees.

Moreover, for developers interested in offering their own AI services through MCP, Docker provides the MCP Toolkit. This toolkit enables users to publish their models for any environment that supports Docker. A dedicated gateway will connect clients, while a specialized terminal facilitates the building, running, and management of services.

Commands for this process include:

• docker mcp server build
• docker mcp catalog add
• docker mcp auth connect

In addition to simplifying the sharing of AI model contexts, Docker emphasizes the importance of security. Service providers can undergo a verification process to receive a blue checkmark, denoting their legitimacy. Authentication and secrets management are streamlined through Docker accounts using OAuth or custom tokens, ensuring that the tools operate securely within their containers.