After many outages we had to build our own free runner https://github.com/supercorp-ai/superci
Hello HN!Code just became cheaper with AI but review and verification is as hard as it was. Tests are great, but current AI slop needs much more exhaustive testing.I built a test coverage tool for coding agents so they could work in a loop and keep writing tests where it makes sense.Here’s Supercov:- Wraps any JavaScript/TypeScript/Rust test suite (npx supercov -- npm test)- Provides deepest...
Built a library to translate between OpenAI Responses/Assistants APIs and other provider APIs. Provides full compatibility so it’s a total drop-in regardless of which provider you use or which features (computer use, web search).End goal will be translating between any SDK and any other SDK with full features, no proxies, no latency.
At first I was very enthusiastic and thought Codex is helping me multiplex myself. But you actually spend so much time trying to explain Codex the most obvious things and it gets them wrong all the time in some kind of nuanced way that in the end you spend more time doing things via Codex than by hand.
So I also dialed back Codex usage and got back to doing many more things by hand again because its just so much faster and much more predictable time-wise.
Guess there’s gonna be more competition haha (but don’t worry, I think our approaches are a bit different).
For hosted MCPs: https://supermachine.ai
https://github.com/supercorp-ai/supergateway
Yo,we’ve just released Supergateway.Now most of MCP servers only support stdio transport and this is problematic if you want remote assistants to access them. MCP spec has SSE transport which works with remote HTTP connections but it is rarely used by MCP server authors. So I’ve created this gateway that allows you to run stdio transport MCP servers and tunnel them via SSE transport.It’s very...
https://github.com/supercorp-ai/supergateway
Hi guys,I’ve just released Supergateway and it allows to use MCP servers with stdio on remote assistants over SSE.Now most of MCP servers only support stdio transport and this is problematic if you want remote assistants to access them. MCP spec has SSE transport which works with remote HTTP connections but it is rarely used by MCP server authors. So I’ve created this gateway that allows you to...
Got any ideas on how to improve it?
https://github.com/Nedomas/databound-rails/commit/04e28afaf0...
For other readers: the issue is being adressed at https://github.com/Nedomas/databound/issues/2
Many Rails apps get/create/update database data on the Javascript side. Sometimes it is done using some kind of framework (Ember, Angular).
This is for the times you don't use a full fledged framework - think React or something similar.
Its benefit is that you don't have to set up routes, parsing/encoding for the data you send.
It lets you securely choose which parts of the model you want the Javascript to edit (permitted_columns).
It also works with Active Model Serializers (f.e. UserSerializer) - it lets you specify which attributes to send to the client.
It is as secure as you would do it yourself.
If possible, please hack the databound.me website and help me fix the holes if there are any.
Thanks!
It all started with a belief that writing unit tests is a much more deterministic process than writing app code. And it still holds true.
Running it would LOCK the class API in tests as it is. So all we would have to do is writing code, locking it, writing more code, locking it too and so on.
I am very thankful that it got featured in the Ruby Weekly today and would bake a cake for the anyone who made it happen. Thanks
Do you think its possible to build writer in such a way so we don't have to write tests anymore?
Running it would LOCK the class API in tests as it is. So all we would have to do is writing code, locking it, writing more code, locking it too and so on.