Model Fitness

Measure which capability tier a model can be trusted with and where its reasoning budget stops paying, then write it into the orchestration roster. Use when a new or unfamiliar model is added, when deciding "can this smaller model handle X", when a delegated agent keeps failing and you need to know whether it is the model or the brief, when choosing a reasoning or effort level, or when setting up tiered delegation for the first time and the roster is empty. Runs a short probe battery and produces a tier ceiling, a budget plateau, and a roster entry.

batteryshark Updated 2 repo stars
Skill Tap — a retro comic-book mascot

File contents

batteryshark/skill-tap commit 7e83ff74a9

Frequently asked questions

npx get-skill add batteryshark/model-fitness