Using Context7 MCP servers for better output
Context7 is an incredible piece of engineering. It's an MCP server that can provide your models with up-to-date documentation of a lot of different libraries and frameworks.
I was working on documenting a fairly large event-sourced application in EventCatalog, and while I had a bit of stuff already documented, I knew documenting a lot of these things would take a lot of time. I've been using Amazon Q Developer with Claude Sonnet 4 a lot lately, experimenting with different MCP servers and offloading some of the boilerplate tasks to an LLM. I'm not sure if it's making me more or less productive, but MCP servers have really done something to the power the tools can provide. Enter Context7.
What has always bugged me is the amount of hallucination some of these LLM's are doing, along with how out of date some of that information is. It looks like Context7 might help me solve that.
Having MCP servers in general feels like a new kind of super power, but having output that actually conforms to the latest version is brand new to me. Being able to ask the model to document a project across 12 repositories, with domains, services, events, commands, queries and the mapping of producers and consumers, and then just have it do its thing with some guidance is a lot faster than doing it by hand.
I'm still 100% sure we shouldn't be vibe coding apps to production, but some of these things do become a lot easier to offload to an LLM to do.