cd /news/ai-tools/make-claude-code-sufferable-with-thi… · home topics ai-tools article
[ARTICLE · art-98090] src=jerodsanto.net ↗ pub= topic=ai-tools verified=true sentiment=↑ positive

Make Claude Code sufferable with this one easy trick

Developer Jerod Santo reports that instructing Anthropic's Claude Code to communicate in ASD-STE100 Simplified Technical English, a controlled language standard from the 1970s, eliminates the AI coding assistant's verbose and clichéd output. Santo, who had previously criticized Claude's writing style, says the one-line memory command makes Claude Code's reports concise and less irritating, reducing mental fatigue during coding sessions.

read2 min views1 publishedAug 15, 2026
Make Claude Code sufferable with this one easy trick
Image: Jerodsanto (auto-discovered)

Make Claude Code sufferable with this one easy trick With every auto-update, Claude Code has gotten better at writing software. What it hasn’t improved at, tragically, is communicating to me in a way that doesn’t make me want to punch it square in the /artifacts

As I’ve written before, Claude is finding its voice and I’m really happy for this important step in its life but also I hate its voice and I hope its voice dies in a fire. You know what I’m talking about. Monstrosities like this:

It’s not an edge case, it’s an invariant violation quietly hiding at the boundary between individually reasonable assumptions, which is exactly why it matters.

[1] I want Claude to keep writing software for me, but I’m done with:

  • “smoking gun”
  • “load baring”
  • “honest opinion”
  • “just say the word”
  • “I have the full picture”
  • “you’re right to push back”
  • “shapes the entire conversation”

Thank God, I recently stopped scrolling 2 long enough to find this

one easy trickthat Andrew Carr

postedon X: I had never even heard of ASD-STE100 before, but like most cool things in our industry, it goes back to the ’70s. Here’s all I needed to know:

The STE standard consists of a set of writing rules and a controlled dictionary. The writing rules focus on grammar and style. The dictionary contains the approved words that a writer can use and a list of words that are not approved with related alternative suggestions.

These approved words were selected because they were simple and easy to recognize. In general, each word has only one meaning and is approved with only one part of speech.

With that knowledge in hand, I fired up claude

and typed:

Make a memory to only report to me in ASD-STE100 Simplified Technical English

You’re curious if it worked. My honest opinion? I was right to push back…

I’ll spare you the Instagram influencer style before-and-after shots. You already have the one easy trick copied to your clipboard.

All you have to do is go paste it.

I yoinked that particular quote from

an Avy Faingezicht essay, because it triggered me, but examples are pervasive↩︎ - Two observed side effects of agentic coding vs hand typing every statement like the bad ol’ days: 1. My brain hurts less at the end of each session. 2. I find myself surfing and scrolling the internet more often.

↩︎

── more in #ai-tools 4 stories · sorted by recency
── more on @jerod santo 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/make-claude-code-suf…] indexed:0 read:2min 2026-08-15 ·