Urgent.News

What's breaking now, across thousands of outlets.

AI

I made my coding agent explain every change to the customer. Here is the skill.

The diff is never the expensive part. The expensive part comes after: explaining to the person who asked for the change what they actually got, and discovering that "done" meant "tests pass". I spent four years as a founding engineer on a hospitality platform and then alone on a proactive assistant, and the habits that saved me the most time had nothing to do with code. Restate the request as a…

Coding agents typically handle change requests, but the real value comes from explaining the outcome to the customer and ensuring the changes behave as expected. The author spent four years building these habits before realizing they were crucial. They now bundle these habits into a skillset called "product-engineer," which comprises seven rules.

The key habits include restating the customer's request, explaining the outcome in plain language, ending commits with a customer-focused block, and confirming what was deliberately not built. The skill also mandates naming what was not shipped, using precise numbers, and speaking the stakeholder's language. There are now four versions of this skill: one for product engineers, one for customers, one for release notes, and one for AI agents that write about software.

Installing the skill is straightforward, and it can be used as a plugin in various coding agents. The skill enforces the rules through a commit message check, ensuring any commit adheres to the seven rules. The author tested the skill across multiple coding tasks and found that using the skill significantly improved the quality of the customer block, what was built, and what was not built.

While the skill does make coding tasks take slightly longer, it ensures a higher quality outcome and better communication with the customer.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Sam Altman says OpenAI going public in 2026 would be ‘ill-advised’

OpenAI CEO Sam Altman confirmed that there would be no OpenAI IPO in 2026 during an interview with Fortune. Over the course of 45 minutes, Altman discussed a variety of subjects including the Hugging Face hacking incident, recursive self-improvement, and the possibility of building an AI that was beyond human control.

I wrote a linter for AI-writing tells. It gave one of my own READMEs a C.

A pull request description came back from my coding agent and I read it the way the reviewer would. An em dash in the first sentence. A bold label on every bullet. Let me know if you need anything else at the bottom. I deleted all of it by hand, the same pass as the day before. Readers have learned these tells.

  • Developed linter to detect AI-generated writing flaws
  • README file received failing grade of F
  • Rules based on Wikipedia guidelines and custom house rules

Anthropic boss Dario Amodei calls for AI development to slow down

The head of AI company Anthropic has called for the pace of development of artificial intelligence models to slow down and to be closely monitored. In an essay on Saturday, Dario Amodei said developing AI was not in question, but the risks associated with it were “serious” and companies and governments must be given time [...] The post…

More from Saturday 12 September →