---
title: "💭 AIs aren’t good rule followers"
description: "!https://x.com/unclebobmartin/status/2044065822067282396"
date: 2026-04-14
published: true
tags:
  - llm
  - ai
  - thought
template: link
---


<div class="embed-card embed-card-external">
  <a href="https://x.com/unclebobmartin/status/2044065822067282396" class="embed-card-link" target="_blank" rel="noopener noreferrer">
    <div class="embed-card-content">
      <div class="embed-card-title">External Link</div>
      <div class="embed-card-meta">X (formerly Twitter) &middot; x.com</div>
    </div>
  </a>
</div>


I've gotta agree with bob on this one, the first thing I did to my biggest brownfield project I wanted to use agents on BEFORE they did work was a hardened pre-commit.yaml, ci, hardened type checking and linting. SECOND get rid of bad inconsistent patterns, let them replicate consistency, force them to pass checks.  Agents will follow all of your markdown suggestions _most_ of the time, enough for you to become complacent if you let it.  They are goal seeking, if you put them to a task you thought was possible that is not given your constraints, they will try to find a way given enough tokens.  I dont see this ever changing, its one thing that makes them great, it just needs to be kept in check.

!!! note

    This post is a <a href="/thoughts/" class="wikilink" data-title="Thoughts" data-description="These are generally my thoughts on a web page or some sort of url, except a rare few don&#39;t have a link. These are dual published off of my..." data-date="2024-04-01">thought</a>. It's a short note that I make
    about someone else's content online <a href="/tags/thoughts/" class="hashtag-tag" data-tag="thoughts" data-count=2 data-reading-time=3 data-reading-time-text="3 minutes">#thoughts</a>
