Saving tokens and "Let me try something else" infinite loop in Coding Agents

With token prices going up for many of the smarter models, I was pondering about how to be more efficient with prompts. Also, there IS an infinite loop possibility in coding agents.

So, knowing what I want was usually the most accurate predictor for costs (other than the small tricks) amongst other things. Here's an outline of some of those solutions about optimizing Claude tokens on my blog