Prompts.
Copy. Run in ChatGPT or Claude with one click. Stolen and refined from issues you should be reading.
The Model Picker
Scorecards disagree, so test on your own work. This prompt interviews you one question at a time about what you use AI for, what you pay now and what matters most, then builds a short, fair test you can run on two or three models in an afternoon.
The Agent Access Check
Decide whether it's actually safe to connect an AI agent like Muse, ChatGPT agent, or Claude to your email, calendar, or payments -- before you do it, not after.
The Longevity Claim Check
The Bottle-to-Dish Matcher
Works out which bottle you already own belongs on tonight's dish, and which one you'd be wasting. Built for balsamic, but works for olive oil, salt, honey, or anything else you keep several versions of.
The AI Video Budget Fit Checker
Figure out which AI video model actually fits your budget once retries and audio are counted — not the per-second sticker price.
The Assurance Decoder
Separates the AI safety and security claims you can actually verify from the ones you are taking on trust - then names the single unverifiable claim that is carrying the most weight in your decisions.
The Delegation Auditor
Finds the places where an automated system could do something you would have said no to. Interviews you about one setup, then ranks the real gaps by consequence rather than by how easy they are to fix.
The Spec Adherence Test
A copy-paste prompt that builds you a side-by-side AI test with a right answer — so you can see which tool actually follows instructions rather than which one just looks impressive.
