How is the community doing authentication & authorization? We're considering how to support in louie.ai, anticipating a large ecosystem of sites supporting it, yet the original spec had essentially TODO for how to do authentication & authorization bindings...
After recently learning about MCP from HN, I started working on some demos. I jotted down my thoughts about my developer experience and potential room for improvement.
Does this work with GPT too or only with Claude? If it does, where is a relevant example?
Really appreciate the write up. I am one of the authors of MCP and this helps a great deal giving a good overview of where we need to do better.
I was a bit under the water over the last few weeks with talking to people about MCP and just generally a bit overwhelmed about how well it has been received.
I'll take a look at issue88 this week. Thanks so much