Welcome to MCP Hub Blog
Why We Started This Blog
The Model Context Protocol (MCP) is reshaping how AI agents interact with the world — but with that power comes complexity: security risks, governance gaps, and the challenge of running MCP servers predictably at scale.
We built MCP Hub to solve those problems. This blog is where we share what we learn.
What You’ll Find Here
Every Tuesday we publish a new article covering:
- Security: vulnerability classes in MCP servers, real-world attack patterns, how to defend against them
- Protocol: deep dives into MCP spec, edge cases, and implementation gotchas
- Tutorials: step-by-step guides for publishing, certifying, and running MCP servers
- Enterprise: compliance, governance, and policy enforcement for AI infrastructure teams
No marketing. No filler. Just technical content from people who build and run MCP infrastructure every day.
Who This Is For
If you build MCP servers, run them in production, or are responsible for the security of AI-connected infrastructure — this blog is for you.
We write for:
- MCP server developers building tools that plug into Claude, Cursor, and other AI clients
- DevOps and platform engineers integrating MCP into internal tooling
- Security teams evaluating the risk surface of AI-connected systems
- CTOs and tech leads making governance decisions about AI infrastructure
Who We Are
We’re Dani Feijoo and Alfonso Muñoz, co-founders of MCP Hub.
Between us, we have decades of experience in security research, platform engineering, and open-source tooling. MCP Hub started as a tool we needed ourselves — a way to publish MCP servers with real security guarantees, not just a package registry.
Get Started
Explore the MCP Hub platform → to publish and certify your first MCP server.
Co-founder of MCP Hub. Security researcher and software engineer specializing in MCP infrastructure and AI governance.