What the heck is MCP? And why is everybody talking about it?
pieces.app
pieces.app
I’m writing this while supervising Claude desktop as it writes out a backend and frontend for a SaaS tool, all based on a simple concept.
Pasted in 10-ish paragraphs of product design and some rough bullet points about data model, stack, screens, and functionality. Then pressed enter and let it rip.
It’s doing its work by creating directories, running commands (eg “npx create-next-app”) and writing files to the file system.
So far it’s bootstrapped an entire python backend with all the right boilerplate, run create-next-app and filled in a full frontend with dozens of components and every page I asked for. I’m reading every line of code as it’s written. It is surreal watching the app magically appear, file by file, in the vscode sidebar.
Everything it’s doing has been according to best practices as of 2025. In an hour of approving tool usages and typing “Continue” into Claude Desktop, it one-shotted an app that could stand a decent chance at a hackathon.
It didn’t do everything in one shot: I had to ask it to make the test suite and had to ask it to replace mock calls in the frontend with real calls to the backend. It trips over when it generates large files, but happily splits them into smaller ones with the prompt “Continue, but split it into several files”.
Claude makes some really nice choices. This is making me seriously reconsider my 6-week-old Cursor subscription. This was all done using just Claude Desktop on macOS.
Got it from this guide: https://www.youtube.com/watch?v=ly3bed99Dy8
I’ve used the Atlassian MCP to help organize my backlog and consolidate tickets into epics, and to leave comments on tickets based on summaries of commits, and even transition tickets to testing and reassign them.
I’ve used the GitHub MCP to write PR descriptions too.
I think once we start getting hybrid MCP + LSP servers (with debugging support), the editors will get even smarter.
- Model Context Protocol on GitHub: https://github.com/modelcontextprotocol - Docs: https://modelcontextprotocol.io/introduction - Matt Pocock MCP Tutorial: https://www.aihero.dev/model-context-protocol-tutorial - Block's Head of Dev Rel's take on Agentic AI and the MCP Ecosystem (and a bunch of other articles): https://block.github.io/goose/blog/2025/02/17/agentic-ai-mcp...
[1] I get USB as a universal interface, but I have a drawer full of USB-C cables that all look alike. It are very different in their capabilities. I am thinking of throwing out all the cheapies and replacing them with $30 cables so I can charge my phone in less than a week.
that said you can mildly ignore at your own risk cable lengths and get 10 or 15ft cables that can and do charge without any noted impact on speed or charger heat but these too are a crapshoot since no name brand carries these due to them being out of spec (I think 3 and/or 6ft are spec, I forget)
Instead of focusing too much on the analogy aspect (which, to be fair, was a very small part of the article), I tried to include lots of project examples so people could get a better idea of what MCP is, and how you can actually use it.
Appreciate the feedback!
But if it's a structured protocol, how can it be general enough to cover all use cases? (say the syntax of an obscure simulation language)
So MCP makes it easy for LLMs to interact with anything. And there are now hundreds of such servers.