I Started Version-Controlling My AI Agent's Prompts Like Production Code. The First "Improved" Prompt Broke Three Days of Silent Work.

Description and use

<p>I started version-controlling my AI agent's prompts like production code. The first "improved" prompt broke three days of silent work.</p> <p>For months my prompts lived wherever they happened to be written: inline strings in Python scripts, a notes app, one in a README I'd forgotten about, and at least two in my shell history. It worked — right up until it didn't, and when it didn't, I had no way to answer the only question that mattered: <em>what changed?</em></p> <p>This is the system I built after that, the failure that forced it, and the parts I'd skip if I were starting over.</p> <h2>

Problem

<p>I started version-controlling my AI agent's prompts like production code. The first "improved" prompt broke three days of silent work.</p> <p>For months my prompts lived wherever they happened to be written: inline strings in Python scripts, a notes app, one in a README I'd forgotten about, and at least two in my shell history. It worked — right up until it didn't, and when it didn't, I had no way to answer the only question that mattered: <em>what changed?</em></p> <p>This is the system I built after that, the failure that forced it, and the parts I'd skip if I were starting over.</p> <h2>

Outcome

<p>I started version-controlling my AI agent's prompts like production code. The first "improved" prompt broke three days of silent work.</p> <p>For months my prompts lived wherever they happened to be written: inline strings in Python scripts, a notes app, one in a README I'd forgotten about, and at least two in my shell history. It worked — right up until it didn't, and when it didn't, I had no way to answer the only question that mattered: <em>what changed?</em></p> <p>This is the system I built after that, the failure that forced it, and the parts I'd skip if I were starting over.</p> <h2>

Evidence

I Started Version-Controlling My AI Agent's Prompts Like Production Code. The First "Improved" Prompt Broke Three Days of Silent Work.

Source