<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Greg Chetcuti</title>
    <description>The latest articles on DEV Community by Greg Chetcuti (@gregdotca).</description>
    <link>https://dev.to/gregdotca</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F2471498%2Fbc0a9f3e-4510-4dbc-8ae9-91ad8392cd8a.png</url>
      <title>DEV Community: Greg Chetcuti</title>
      <link>https://dev.to/gregdotca</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/gregdotca"/>
    <language>en</language>
    <item>
      <title>AI Code Rating – Three Characters That Say How AI Was Used in Your Project</title>
      <dc:creator>Greg Chetcuti</dc:creator>
      <pubDate>Wed, 11 Nov 2026 18:04:59 +0000</pubDate>
      <link>https://dev.to/gregdotca/ai-code-rating-three-characters-that-say-how-ai-was-used-in-your-project-2hb0</link>
      <guid>https://dev.to/gregdotca/ai-code-rating-three-characters-that-say-how-ai-was-used-in-your-project-2hb0</guid>
      <description>&lt;p&gt;When I'm deciding whether to depend on a project, I'd like to know how it was built. Did a person write this? Did AI write it, with someone reading every change? Or did it go from a prompt to the main branch without anyone looking?&lt;/p&gt;

&lt;p&gt;Today you have to guess from the commit history. So I made &lt;a href="https://aicoderating.com" rel="noopener noreferrer"&gt;AI Code Rating&lt;/a&gt;, a three-character rating that a project publishes in an &lt;code&gt;ACR.md&lt;/code&gt; file at the root of its repository.&lt;/p&gt;

&lt;h2&gt;
  
  
  The Three Positions
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Position&lt;/th&gt;
&lt;th&gt;Range&lt;/th&gt;
&lt;th&gt;What it says&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Maintainer Expertise&lt;/td&gt;
&lt;td&gt;
&lt;code&gt;A&lt;/code&gt;–&lt;code&gt;E&lt;/code&gt;
&lt;/td&gt;
&lt;td&gt;How well the person responsible for the code knows this language and domain (in a solo project that's you; on a team, whoever approves changes), from Expert to Non-programmer&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;AI Share&lt;/td&gt;
&lt;td&gt;
&lt;code&gt;0&lt;/code&gt;–&lt;code&gt;4&lt;/code&gt;
&lt;/td&gt;
&lt;td&gt;How much of the code AI wrote, from none to 1–25%, 26–50%, 51–75%, and 76–100%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Oversight&lt;/td&gt;
&lt;td&gt;
&lt;code&gt;a&lt;/code&gt;–&lt;code&gt;e&lt;/code&gt;
&lt;/td&gt;
&lt;td&gt;How carefully AI changes were checked, from Verified (read line by line and tested) to Unchecked (merged as generated)&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Here are some examples.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;&lt;code&gt;A2b&lt;/code&gt;&lt;/strong&gt; is an expert maintainer, 26–50% of the code written by AI, and every AI change read before merge.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;&lt;code&gt;B0a&lt;/code&gt;&lt;/strong&gt; is an experienced maintainer with no AI-written code. When AI Share is &lt;code&gt;0&lt;/code&gt;, Oversight is always &lt;code&gt;a&lt;/code&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;&lt;code&gt;D4e&lt;/code&gt;&lt;/strong&gt; is a maintainer still learning the stack, nearly all the code written by AI, merged as generated.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Every rating has its own page that explains it, such as &lt;a href="https://aicoderating.com/A2b/" rel="noopener noreferrer"&gt;aicoderating.com/A2b/&lt;/a&gt;. The &lt;a href="https://aicoderating.com/examples/" rel="noopener noreferrer"&gt;examples page&lt;/a&gt; works through common setups, with the reasoning for each position.&lt;/p&gt;

&lt;h2&gt;
  
  
  It Doesn't Judge
&lt;/h2&gt;

&lt;p&gt;AI Share describes how the code was made, not how good it is. An &lt;code&gt;A3a&lt;/code&gt; project, where an expert wrote most of the code with AI and checked every change, can be stronger than a &lt;code&gt;C1e&lt;/code&gt; project with a little AI code that nobody checked. The rating gives readers the facts and lets them decide what matters to them.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why Rate the Maintainer?
&lt;/h2&gt;

&lt;p&gt;Most AI disclosure standards use a single level, such as "none", "assisted", or "generated". That folds several questions into one answer. The question I think gets missed is who's doing the checking. The same AI-written code means something different when someone who could have written it reviews it than when nobody on the project can read it.&lt;/p&gt;

&lt;p&gt;The level is self-assessed against two practical questions. Could the person responsible for the code have written it without AI, and can they explain every line? The &lt;a href="https://aicoderating.com/related-work/" rel="noopener noreferrer"&gt;Related Work&lt;/a&gt; page compares AI Code Rating with AI-DECLARATION.md and other standards.&lt;/p&gt;

&lt;h2&gt;
  
  
  What the File Looks Like
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight markdown"&gt;&lt;code&gt;---
rating: B4b
spec: "0.1"
updated: 2026-11-07
---

&lt;span class="gu"&gt;# AI Code Rating
&lt;/span&gt;
&lt;span class="gs"&gt;**[ACR B4b](https://aicoderating.com/B4b/)**&lt;/span&gt; (Maintainer: Experienced · AI Share: 76–100% · Oversight: Reviewed)

This project is rated with [AI Code Rating](https://aicoderating.com/spec/0.1/), spec version 0.1.

&lt;span class="gu"&gt;## How AI Was Used
&lt;/span&gt;
A few sentences in your own words.
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;That's the actual rating of the AI Code Rating site. Claude Code wrote nearly all of its code and drafted most of its text, and I directed every change and read all of it before it went in. My older project &lt;a href="https://github.com/domainmod/domainmod" rel="noopener noreferrer"&gt;DomainMOD&lt;/a&gt;, written almost entirely by hand since 2010, is &lt;code&gt;B1b&lt;/code&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  Getting Your Rating
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;Answer a few questions in the &lt;a href="https://aicoderating.com/#rate" rel="noopener noreferrer"&gt;rating form&lt;/a&gt;. It writes the &lt;code&gt;ACR.md&lt;/code&gt; file, with the rating linked to its page, and a README badge (Markdown, HTML, or reStructuredText) for you.&lt;/li&gt;
&lt;li&gt;Add the file to the root of your repo.&lt;/li&gt;
&lt;li&gt;Check it with the &lt;a href="https://aicoderating.com/validate/" rel="noopener noreferrer"&gt;validator&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;Optionally add the &lt;a href="https://github.com/ai-code-rating/action" rel="noopener noreferrer"&gt;GitHub Action&lt;/a&gt;, which checks the file in CI and warns if your README badge shows a different rating.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;The badge comes from shields.io, so nothing about your project depends on the AI Code Rating site.&lt;/p&gt;

&lt;h2&gt;
  
  
  Can't People Just Lie?
&lt;/h2&gt;

&lt;p&gt;Yes, ratings are self-reported. But an inflated rating is easy to contradict, since commits, pull requests, and issues show how a project really works, and a rating that doesn't match them costs trust. An honest &lt;code&gt;D4c&lt;/code&gt; tells readers more than a doubtful &lt;code&gt;B4b&lt;/code&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  Get Listed
&lt;/h2&gt;

&lt;p&gt;Once your &lt;code&gt;ACR.md&lt;/code&gt; passes the validator, &lt;a href="https://github.com/ai-code-rating/aicoderating.com/issues/new?template=add-project.yml" rel="noopener noreferrer"&gt;ask to be added&lt;/a&gt; to the &lt;a href="https://aicoderating.com/#rated" rel="noopener noreferrer"&gt;Rated Projects&lt;/a&gt; list on the home page. It's a short form that asks for the project's name, repository, rating, and a one-sentence description.&lt;/p&gt;

&lt;p&gt;The &lt;a href="https://aicoderating.com/spec/" rel="noopener noreferrer"&gt;spec&lt;/a&gt; is version 0.1, under CC BY 4.0, and the site code is MIT. Issues and suggestions are welcome &lt;a href="https://github.com/ai-code-rating/aicoderating.com/issues/new/choose" rel="noopener noreferrer"&gt;on GitHub&lt;/a&gt;.&lt;/p&gt;

</description>
      <category>ai</category>
      <category>opensource</category>
      <category>github</category>
      <category>programming</category>
    </item>
  </channel>
</rss>
