claudish-to-english

It's a shame that I have to run a local model to decipher Opus, but them "dumber" models read far more naturally - https://github.com/gvzdv/claudish-to-english

I had to develop a course related to quantum chemistry and unfortunately, Claude just kept writing it in its own very annoying claude-speak. It actually cost our team a lot of time and tokens to wrestle it to just speak like a normal human. We used skills, as well as initial prompts and subagents to make sure that all of the tics were gone, but there was one too many times where it just wouldn't fix itself despite telling it in the prior prompt! I made speak-english based off of https://github.com/gvzdv/claudish-to-english/commits/main/ and it runs a qwen model on my gcp servers. It can convert documents or respond in-line in sessions with claude and I anticipate building a solution for prime-agent.

Similar to vomit[0] and claudish-to-english[1]. My general feeling here is that this really needs to be fixed at the model level. All other solutions are mere hacks that work for a bit but all have drawbacks. It's either adding yet more context (which has variable rates of success) or piping to a different model to translate, or using a skill, or mentioning this per each prompt, or.... At this point I think it's in Opus's nature to talk like this, and given enough back-and-forth it will start to drift back to its classic mode, aka this incomprehensible "load-bearing seam" babble. [0] https://github.com/zachahn/vomit/tree/main [1] https://github.com/gvzdv/claudish-to-english

Also related and recent: https://news.ycombinator.com/item?id=49265305 "Claudish to English" (https://github.com/gvzdv/claudish-to-english) 4 points | bryan0 |10 days ago | 2 comments

I haven't tried with Claudette but did evaluate https://github.com/zachahn/vomit and https://github.com/gvzdv/claudish-to-english earlier today. I ended up using claudish-to-english with a prompt derived from the vomit one and some of my existing instructions. I ran a few local models through a test harness to see how they did, and the gemma4 ones added bad behavior back in less than any others I tested. So for the parent's Macbook question `gemma4-26b-mlx` should work well. For you with 24 GB VRAM, `gemma4-26b-a4b`. I tried higher VRAM models and they slowed down while still doing just as well or slightly worse. If someone else tests and finds a better performing model though, please update me here, I'd love to try it.

Check out the awesome [Claudish-to-English plugin](https://github.com/gvzdv/claudish-to-english), you could lift their prompt from https://github.com/gvzdv/claudish-to-english/blob/main/rewri...

It is so coincidental that I have been working on a fork of Kilo to do the exact same thing. I wonder if it has any effect on watermarking. I came across this other project yesterday which appears to have the same goal (unaffiliated and I havent tried it). https://github.com/gvzdv/claudish-to-english

So here's the workflow I gathered, from the comments I've read here recently: - Claude as main agent, but use this skill[0] to make Claude delegate everything to Codex, because it's cheaper and faster. (Hilariously, the skill is official!) - Use TFA or Claudish to English[1] so the final output is actually human readable. Ironically it wasn't so long ago that I was asking Claude to rewrite output from other LLMs to make it more readable... [0] - https://github.com/openai/codex-plugin-cc [1] - https://github.com/gvzdv/claudish-to-english

I like the "Claudish to English" name better. https://github.com/gvzdv/claudish-to-english

Wtf, why isn't this the default? It says it can use any model/provider (but recommends Gemma?) I must investigate. This is probably where 80% of my cognitive load comes from these days: the "language barrier". (How ironic!) EDIT: > You rewrite the assistant's message into much simpler, plain English. Keep every fact, name, number, and file path. Use short sentences and everyday words. Leave fenced code blocks unchanged. Output ONLY the rewritten message with no preamble, labels, or commentary. > For context, the user asked the assistant: "$userq". Use this only to understand the message. Do NOT rewrite, answer, or repeat the user's question — rewrite only the assistant's message that follows. https://github.com/gvzdv/claudish-to-english/blob/main/rewri... This is even better than Caveman.

I've been using this CC hook with decent results: https://github.com/gvzdv/claudish-to-english It uses a local LLM to translate Claude's output.