The latest articles on DEV Community by SalzDevs (@salzdevs_0). https://dev.to/salzdevs_0 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%2F4007011%2Fd557c781-b377-451a-a77c-451a86ca382f.png https://dev.to/salzdevs_0 en SalzDevs Sun, 28 Jun 2026 21:37:10 +0000 https://dev.to/salzdevs_0/i-got-tired-of-repeating-the-same-ai-coding-instructions-so-i-built-ai-rules-35o0 https://dev.to/salzdevs_0/i-got-tired-of-repeating-the-same-ai-coding-instructions-so-i-built-ai-rules-35o0 <p>I’ve been using AI coding agents more and more, and one thing kept annoying me:<br> I was repeating the same instructions every session.<br> Things like:</p> <ul> <li>keep diffs small</li> <li>don’t fetch data directly inside React components</li> <li>follow the existing architecture</li> <li>add tests when behavior changes</li> <li>don’t introduce new patterns unless needed</li> <li>explain risky changes before making them At first, I handled this with big instruction files like <code>AGENTS.md</code>. That works for a while, but it has two problems:</li> <li>The file keeps growing.</li> <li>Most of the instructions are irrelevant to the current task. If I’m asking the agent to change a React component, it probably doesn’t need my rules about commit messages, backend handlers, database migrations, or release notes. And if I work across multiple projects, I either copy those instructions around or rewrite them again. So I built <strong>ai-rules</strong>. ## What ai-rules does <code>ai-rules</code> is a small local-first CLI for personal AI coding rules. Rules live as Markdown files in:</li> </ul> <div class="highlight js-code-highlight"> <pre class="highlight plaintext"><code> bash ~/.config/ai-rules/rules/ </code></pre> </div> ai