author, source, and corrections
I built whoburnedmore because the usage data already sits on every developer's machine — Claude Code, Codex, Cursor, Gemini CLI and a dozen other agents each keep their own local logs — but nothing pulled them together into one honest number you could compare. So I made a single command that reads those logs locally and ranks your token burn and cost. Only the daily totals ever leave your machine; never your prompts, code, or file names.
Guides on this site combine product testing, public documentation, and the limitations of the usage formats the project can read. They are reviewed against real output where the project has a supported reader, rather than presented as universal vendor truth. If something here is wrong or out of date, the source is open and corrections are welcome.
One command reads the logs your AI coding tools already keep and puts you on the leaderboard.