jbentley1

You can use Crystal (https://github.com/stravu/crystal) to run Codex and Claude Code at the same time and just pick the best result.

We built a tool to use Git Worktrees to manage all of our in-progress agents. It's expanded from just Claude Code to having Codex support, and I'm loving running them side by side. They see things from different angles, make different UX choices, find bugs the other one doesn't, etc.

Just having the model variety is helpful, it is my default way to work now for every prompt I submit.

Check out Crystal, similar but open source https://github.com/stravu/crystal

I do this every day! In fact, I built a new type of IDE around this (https://github.com/stravu/crystal) and I can never go back.

I've built something very similar but am further along (https://github.com/stravu/crystal). I understand the desire to do your own thing, but if you are interested in joining forces and contributing I would love to have you. I think we were thinking along very similar lines.

I use Crystal which archives all my old claude code conversations, I've had to do this a few times when I threw out code that I later realized I needed.

I'm going to use this for all of my Slack replies

A new type of development environment for working with agents

https://github.com/stravu/crystal

It supports Claude Code and Codex, but has you constantly working on multiple features in Git worktrees. This way you are always able to stay busy while waiting on your agents.

It has built in tools for review, such as a diff viewer, and a quick button to run your application in different worktrees for testing. It has completely transformed the way I work.

Yeah I didn't realize there was also a Ruby dialect called Crystal until we had already launched. Not that I could think of a better name anyways.

We've added support for Codex in Crystal, meaning you can now run Codex sessions alongside Claude Code in isolated git worktrees for parallelized development.

This is a great list for people who want to smugly say "Um, actually" a lot in conversation.

Based on my brief stint doing data work in psychology research, amongst many other problems they are AWFUL at stats. And it isn't a skill issue as much as a cultural one. They teach it wrong and have a "well, everybody else does it" attitude towards p-hacking and other statistical malpractice.

I am by no means an expert on this, but was reading this about the amoloid hypothesis this morning: https://www.astralcodexten.com/p/in-defense-of-the-amyloid-h...

He addresses the specific reasons that the mouse results don't translate, but the interesting part to me is that the scientific community seems to be behind the amyloid hypothesis while the media likes to run with any story that refutes it.

It's fully open source, you can manage Git Worktrees, review diffs, run your application, and push your branches all in one place. It is a re-imagining of what an IDE should be in 2025.

I originally built this for myself because I didn't like waiting around for Claude Code to finish and I didn't want to mess with a bunch of terminal windows to use worktrees so I built a GUI. Then I started adding more features, diff view, script running, and more and my team and I decided it needed to be shared as an open source project.

Excited to hear what the HN community thinks.

I honestly think all of these VSCode forks are moving in the wrong direction. They are horseless carriages, old-style tools with AI stapled to the side. I'm trying to figure out what agentic coding should look like, and I built Crystal as an agent management platform first, IDE second.

https://github.com/stravu/crystal/

I built Crystal to manage Claude Code agents in worktrees instead of using an IDE that only lets me work on one thing at a time.

Naturally skinny people hate Ozempic for the same reason I know many talented engineers who hate AI coding.

This is great. I hope that they expose the configuration through the SDK, but right now it works with Crystal (https://github.com/stravu/crystal) if you configure them in the CLI.

I'm working on a proper config screen for them that just modifies the agent files directly, and a future release will also give special formatting for agent output.

Crystal is similar and works with existing checked out repos https://github.com/stravu/crystal

Crystal can do all of this and more, and unlike Conductor is open source.

https://github.com/stravu/crystal

My Claude Code usage would have been $24k last month if I didn't have a max plan, at least according to Claude-Monitor.

I've been using a tool I developed (https://github.com/stravu/crystal) to run several sessions in parallel. Sometimes I will run the same prompt multiple times and pick the winner, or sometimes I'll be working on multiple features at once, reviewing and testing one while waiting on the others.

Basically, with the right tooling you can burn tokens incredibly fast while still receiving a ton of value from them.

https://github.com/stravu/crystal

Crystal is a re-imagining of what an IDE means when AI drives development. Traditional IDEs are designed for deep focus on one task at a time, but that falls apart when you have to wait 10-20 minutes for an agentic task to run. Crystal lets you manage multiple instances of Claude Code so you can inspect/test the results of one while waiting for the others to finish.

Is this the same as the prompt caching that other API's (Anthropc, OpenAI, etc) have had, just open source and for vLLM?

If you can call Claude Code that means everything else should be working, as most functionality is built around the terminal and that is how it is calling Claude Code.

Thanks for your help, now I'll be able to include Linux support in my next release

While I seem to have a little attention from this comment, if anyone can test this Linux installer for Crystal and tell me if it works on their machine I would appreciate it:

https://github.com/stravu/crystal/actions/runs/15791009893/a...

If you are paying a senior engineer 200k, getting them a CC max plan is equivalent to 1.2% of their salary. I would say that it increases productivity by a lot more than that.

So yes it might feel expensive in terms of a personal monthly budget, but the value for money is insane.

Weird, I have not had this issue and I commonly run 5+ at once

If I hit the rate limit in 2 hours and got value out of each prompt I ran, that's better than doing the same amount of work in 6 hours and not hitting the limit.

Personally, I'm running 2 accounts and switching between them for maximum productivity. Just as a function of what my time is worth it is a no brainer.

The other replies are correct that right now you need some level of human review, but it would be interesting to have a second AI review with a clean context. Maybe a security checklist, or a prompt telling it to check that the tests are covering the functionality appropriately.

I tried to make it easy to remember what you are doing. You can see the prompts you ran, and I used the Monaco editor from VSCode to view and edit the diffs.

I think there are opportunities to give special handling to the markdown docs and diagrams Claude likes to make a long the way to help review.