See why your AI agents fail, and what they cost.

Earlyman reads your agent traces on your Mac, names the step that broke the run, and shows what your agents and models cost.

Free and open source. No account. Runs on your Mac.

EarlymanSample data
Pull request #214branch feat/checkout-retry
Cost$3.84 estimate
Tokens2.4M 412 requests
Sessions3
Agent time41 m
Failed tool calls6 4 recovered
Issues2 fixed with proof
Rules1 broken
How it works

From one agent step to the monthly bill.

Earlyman never stores or sends your prompts.

It keeps counts, not content.