MCP turned agent tooling into a protocol problem, and the ecosystem around it is developing fast — servers, transports, discovery surfaces, and a lot of unsettled convention.
These posts are field notes from building and operating an MCP server: transport choices, security hardening, tool design, and where the ecosystem seems to be heading.
The moment an MCP server leaves stdio for HTTP, it becomes a network service with everything that implies. The hardening playbook I actually run.
Every middleman reselling API calls adds margin and a second data controller. The case for bring-your-own-key as trust architecture, tradeoffs included.
One transport is a subprocess and a pipe. The other is a service you now operate. Which MCP transport you pick is really a question about who connects, from where.
Forget the spec tour. This is what MCP looks like from the inside of a server that runs every day: the parts that stuck and the parts that didn't.