Live data from Hacker News

MCP Gateway and Registry

github.com

41–50 of 58 posts

Re: MCP Gateway and Registry

#41
I have built similar gateway but as a desktop application which runs locally and gives you a lot of control over enabled MCP tools and allows you to have a separate profiles for different combinations of MCP servers and credentials (e.g. profile with production API keys, sandbox keys and etc,.).

give it a try here: https://getaiko.app/downloads

Re: MCP Gateway and Registry

#42
Ah nice an AI generated MCP gateway. I can't wait to see how many different agent/tool protocols we'll run through this year, and how many gateways. I just spin up oauth2proxy to handle authentication with istio gateway. Maybe I should be deploying piles of unknown code instead?

Re: MCP Gateway and Registry

#43

Does anyone use MCP within their product, or is the MCP ecosystem entirely focused on enabling new workflows within an existing chat/agent framework (a la Cursor or Claude code)? Assuming the latter, are there any viable non-developer MCP clients?

I was really hoping this would take off. https://github.com/universal-tool-calling-protocol

I don’t really get it — designing a truly universal description language for all possible service endpoint types and protocols is a hard (arguably impossible) problem.

Writing a client capable of parsing and using that description isn’t much easier.

Unsurprisingly, it appears there’s no real spec: https://www.utcp.io/about/RFC

Re: MCP Gateway and Registry

#44
post #16
post #11

I built hyper-mcp[1] as side project with each MCP as a WASM plugin. The idea is each plugin runs in their own wasm vm with limited network/file system access. Plugins are written in any language, as long as they can compile to WASM and publish to OCI registry (signed & verified with sigstore) Recently, Microsoft released their own version of hyper-mcp named Wassete[2] Ideally, I want to make it like a gateway with m…

Interesting we also just released our mcp gateway project under the name Hypr MCP[1]. [1]: https://github.com/hyprmcp/mcp-gateway

Cool. I want to add dynamic client registration but haven't got around to do it yet.

Re: MCP Gateway and Registry

#45

Ah nice an AI generated MCP gateway. I can't wait to see how many different agent/tool protocols we'll run through this year, and how many gateways. I just spin up oauth2proxy to handle authentication with istio gateway. Maybe I should be deploying piles of unknown code instead?

oauth2proxy actually doesn't work for MCP servers due to missing dynamic client registration..

Re: MCP Gateway and Registry

#46
post #44
post #16

Earlier quoted context omitted.

Interesting we also just released our mcp gateway project under the name Hypr MCP[1]. [1]: https://github.com/hyprmcp/mcp-gateway

Cool. I want to add dynamic client registration but haven't got around to do it yet.

Feel free to just include our gateway in your docker compose file.

Re: MCP Gateway and Registry

#47
post #45

Ah nice an AI generated MCP gateway. I can't wait to see how many different agent/tool protocols we'll run through this year, and how many gateways. I just spin up oauth2proxy to handle authentication with istio gateway. Maybe I should be deploying piles of unknown code instead?

oauth2proxy actually doesn't work for MCP servers due to missing dynamic client registration..

You're right, we're not strictly implementing things according to the spec/oauth 2.1

Re: MCP Gateway and Registry

#48

Earlier quoted context omitted.

I was really hoping this would take off. https://github.com/universal-tool-calling-protocol

I don’t really get it — designing a truly universal description language for all possible service endpoint types and protocols is a hard (arguably impossible) problem. Writing a client capable of parsing and using that description isn’t much easier. Unsurprisingly, it appears there’s no real spec: https://www.utcp.io/about/RFC

> designing a truly universal description language for all possible service endpoint types and protocols is a hard (arguably impossible) problem

Fair enough, but when restricting yourself to the problem of tool calling, giving the AI the ability to call APIs in a simplified way, rather than inventing a new communication protocol like MCP, is, in a lot of cases, better.

Post reply on HN