@andrewchemis· 4h
Agent debugging of andrewchemis.dev
andrewchemis.dev
FULL SEND
168.4x your typical session
3.2M tokens vs typical 19.0K
3.2Mtokens
15prompts
4:23:18time
216lines
Claude Codeclaude-fable-5-1Markdown$66.67
mostly delegating
Autonomous 8h 56m agent session running commands and testing. 568K tokens processed using claude-fable-5-1.
Tokens / prompt213.5k
Cost / line$0.309
Cache hit95%
Burn rate$15.19/hr
Derived from this session's token and cost data. Not shown on the feed.
Building 0%Verifying 4%Thinking 1%Delegating 95%
Comments