Ab

Run an A/B test of the same task at multiple model tiers (Haiku, Sonnet, optionally Opus and Fable). Captures outputs, computes structural and semantic diffs, scores quality, writes a markdown comparison report. Use when the user wants to validate "is Haiku good enough for this task class?" or runs /tokenwise:ab "<task description>".

CodeShuX Updated
TokenWise — Cut your Claude Code spend, measured on your real work, not a marketing claim.

File contents

CodeShuX/tokenwise commit 73ff99d8b6

Frequently asked questions

npx get-skill add codeshux/ab