I’m interested in these types of tools, but the biggest issue I am currently having in my workflow is manually verifying everything is working. Tests and stuff are nice, but typically if there is a bug, the AI agents enshrine it in th tests.

However with worktrees I am not really able to easily copy secrets, etc to run my app, ports conflict, I end up with a bunch of separate dbs and services, etc.

Does conductor help with this? Have you all found any useful ways of making this easier or more automated?

Conductor has a feature that lets you define a script to copy .env and other things as needed from main repo into new worktrees on creation.

I don't use it much anymore, but last I did I would run into port conflicts with docker projects.

Qwen is probably your best bet…

Edit: I’d also consider waiting for WWDC, they are supposed to be launching the new Mac Studio, an even if you don’t get it, you might be able to snag older models for cheaper

> consider waiting for WWDC

100% agree. I’m just looking forward to setting something up in my electronic closet that I can remote to instead of having everything tracked.

Latest rumors are no Mac Studio until at least October.

They left openAI for ideological safety reasons, if you believe their corporate lore.

They present themselves as an org with some ideology

.net will work. Use a weaver (fody) or the modern features to roll everything into 1 .exe.

Use self-contained to have everything together.

https://learn.microsoft.com/en-us/dotnet/core/deploying/sing...

Core is compatible with non deprecated apis.

That’s why they had .NET 5im stead of .net core 5

1 Michelin star is like _only_ in the top 0.1% of restaurants instead of the top 0.001%.

It’s still impressive, difficult, and time consuming.

Highly recommend you check out any starred restaurants nearby where you live. They tend to be expensive, but they are worth the high sticker price

I don't think you caught the sarcasm of parent

Copilot feels like being a caveman, Claude code feels like modern times comparatively.

No, those are the models. They are NOT the same as Claude code/codex.

GitHub copilot agent mode is quite bad, and advanced premium models like opus4.6/codex 5.3 are making it barely usable.

Claude code and codex are night and day comparatively.

Additionally, GitHub copilot admins can enable/disable models. Each model has different ToS, so perhaps their admin has not turned it on.

I would suggest you not immediately insult people with ad hominem attacks.

Copliot is quite bad out of the box, but it does work with opencode, which I found to be as good a harness as claude code, and a better tool overall.

Wait really? I thought the models were basically the same with a wrapper.

What about warp, is that also just a wrapper?

There's a layer between your input and the model called an agent harness. It's the bit that guides the model how to traverse the file system, where to search, how the codebase is architected, how to navigate the monorepo.

When you say "Add a default $5 tip to the dialog screen titled 'Tip this waiter?'", what the harness does is supply information on where the strings are, the dialogs, and where the design style containing the PrimaryButton might be.

Cursor is excellent at this and probably pioneered the whole approach. Copilot hasn't really bothered to be more than a wrapper.

The models are the same. The agent implementation is different. I can confirm Claude Code performs much better than GH Copilot with the same Claude models.

Orion is Kagis browser.

Atlas is OpenAIs browser