1
Ask HN: How are you using Claude Code without blowing through tokens?
I’ve been using Claude Code for my personal projects and I like it, but I feel like I’m probably using it inefficiently.
Right now I mostly have it work off markdown files for project context/rules, plus get-shit-done and frontend skill occasionally. It works, but token usage climbs fast.
How are people actually using it without wasting tokens?
Are you keeping instructions tiny, using cheaper models, implementing specific skills/plugins, or doing something smarter?
What made the biggest difference for you in real projects?