websocket VS go-rabbitmq-example

What I can say is that if you will use Gorilla Websocket (https://github.com/gorilla/websocket) (it's archived now unfortunately, but still super robust) - then you can go to sth like 15-20kb per connection - see https://github.com/FZambia/go_websocket_memory - to achieve this you need to reduce read buffer size, reuse write buffers, use goroutine after hijacking connection (following chat example in Gorilla WebSocket repo).

No Bugs? https://github.com/gorilla/websocket/issues Even if that were true at the time of stoping development, "No bugs" is something that can never be said for any software. I really wish this community stopped recomending to use abandoned libraries if only for the potential Security issues. A not maintained Library is a very lucrative target for vulnerability hunting.

The Gorilla WebSocket License says:

We'll be using WebSocket library: "github.com/gorilla/websocket" ❣️

package github.com/gorilla/websocket is not a main package

As for Websockets, (https://github.com/gorilla/websocket) it uses ws.NextReader() to handle heavy loads of connections.

I'm looking to build a game server that uses web sockets to communicate with clients, I've looked into some of the libraries and the main ones seem to be gorilla websockets and melody.

If this is what your import looks like "github.com/gorilla/websocket" that's wrong it should be just "github.com/gorilla/websocket"

Most of the tutorials that I see to use rabbitmq with go are using amps (like this). That is all fine and dandy, but I still could not find any tutorial on how to integrate the amps consumer output to http handler? I guess I should get the consumer output through a middleware but have difficulty on how to do so.