Skills, Hooks & Making Claude Yoursexpert · ~165 min

The Skill That Quietly Stopped Firing — Build a Routing Regression Suite for Your Whole .claude Pack

Your skills fired reliably the week you wrote them. Then you added nine more, a teammate added two, and a release changed how much room the skill list gets. Nothing in your repository changed. Can you prove — today, with evidence you could hand to someone else — that each of your skills and hooks still fires on the requests it was built for, and that it is your customization producing the result rather than a capable model that would have done the same thing anyway?

🤝 Want this built for you? Start a request — our AI scopes it, estimates a price range, and matches you with a vetted AI Advisor. →

Two ways to open it

Just this build

$10 one-time

Unlocks this full walkthrough on your account — no subscription, and no account needed until you check out.

Best value Over 150 projects + the full course

$99 / year

Over 150 project walkthroughs, the 6-week course, and your certificate.

🔒 Secure checkout by Stripe · Already a member? Sign in

What you'll do

A headless routing regression suite for your whole .claude pack that runs each probe several times across a pack arm and a control arm, reports pass, fail, or inconclusive with a fire rate rather than a single boolean, names the specific cause class behind a failure instead of only reporting a red line, keeps the raw stream evidence for every run, quarantines the probes that are genuinely too noisy to gate on, and runs before and after every Claude Code upgrade so a silent routing regression has somewhere to be caught.