Today, we want to highlight Docker MCP Toolkit, a free feature in Docker Desktop that gives you access to more than 200 MCP servers. It’s the easiest and most secure way to run MCP servers locally for your AI agents and workflows. The MCP toolkit allows you to isolate MCP servers in containers, securely configure individual servers, environment variables, API keys, and other secrets, and provides security checks both for tool calls and the resulting outputs. Let’s look at a few examples to see it in action.
Get started in seconds: Explore 200+ curated MCP servers and launch them with a single click
Docker MCP Catalog includes hundreds of curated MCP servers for development, automation, deployment, productivity, and data analysis.
You can enable MCP servers and configure them with just a few clicks right in Docker Desktop. And on top of that automatically configure your AI assistants like Goose, LM Studio, or Claude Desktop and more to use MCP Toolkit too.
Here are two examples where we configure Obsidian, GitHub, and Docker Hub MCP servers from Docker MCP Toolkit to work in LM Studio and Claude Desktop.
Build advanced MCP workflows: Connect customer feedback in Notion directly to GitHub Issues
And you can of course enable setups for more complex workflows involving data analysis. In the video below, we use Docker Compose to declaratively configure MCP servers through the MCP Gateway, connected to the MCP Toolkit in Docker Desktop. The demo shows integrations with Notion, GitHub MCP servers, and our sample coding assistant, Crush by Charmbracelet.
We instruct it to inspect Notion for Customer Feedback information and summarize feature requests as issues on GitHub. Which is a nice little example of AI helping with your essential developer workflows.
Learn more about setting up your own custom MCP servers
And of course, you can add your custom MCP servers to the MCP Toolkit or mcp-gateway based setups. Check out this more involved video.
Or read this insightful article about building custom Node.js sandbox MCP server (article) and plugging it into a coding agent powered by one of the world’s fastest inference engine by Cerebras.
Conclusion
The Docker MCP Catalog and Toolkit bring MCP servers to your local dev setup, making it easy and secure to supercharge AI agents and coding assistants. With access to 200+ servers in the MCP Catalog, you can securely connect tools like Claude, LM Studio, Goose, and more, just a few clicks away in MCP Toolkit. Check out the video above for inspiration to start building your own MCP workflows! Download or open Docker Desktop today, then click MCP Toolkit to get started!
Learn more
- Explore the MCP Catalog: Discover containerized, security-hardened MCP servers
- Download Docker Desktop to get started with the MCP Toolkit: Run MCP servers easily and securely
- Check out our MCP Horror Stories series to see common MCP security pitfalls and how you can avoid them