Motivation Some months ago I realised I was the bottleneck. I am someone who still likes to read the plans and outputs the LLMs create. When working on multiple projects at a time, this starts to pile up and the rate of approval and understanding drops a lot. To add to that, I discovered the tools I was using to review the plans and outputs of the LLM were also flawed, and contributed to the problem. ...
Every week someone posts a screenshot of an agent deleting something it shouldn’t have, and the replies underneath say the same thing: the model is reckless and shouldn’t need babysitting. I understand the feeling and I think it’s a fair one. These things have improved so fast that expecting them not to make silly mistakes has started to feel reasonable. They still make them. But the distance between what a model might do and what it can actually reach on your machine is a config file most people have never opened. You need permissions to keep things running smooth, not faith. ...