Debugged fielddaycommander
1.6M tokens vs typical 209.6K
1h 57m session running commands and testing. 508K tokens across 5 prompts using claude-opus-5-5.
Derived from this session's token and cost data. Not shown on the feed.
Comments