The real problem isn't cooking, it's deciding

I used to stand in front of the fridge at 6:30pm most nights with zero idea what I was making, not because I couldn't cook, but because deciding was the one task I had no energy left for. By the time dinner rolls around, you've already made a hundred small decisions that day. Food shouldn't be the hundred-and-first.

That's the actual job here — not recipes, not nutrition science, just removing one recurring decision from your day before it costs you money and wasted groceries too.

What I'd actually hand off

I run three of our own agents for this, and they split cleanly into three different jobs instead of one vague "meal planning" blob:

• MealPlannerSupreme decides the what — a week of meals built around your actual preferences, not a generic diet template. Saves roughly 5 hours a week once you've told it your constraints once.
• GroceryGenius turns that into a list sorted by what you'd actually buy, and skips anything already sitting in your pantry. Another ~5 hours a week, mostly the hunting-through-the-kitchen-first kind of hours.
• RecipeTransformer handles the nights where you've got three things in the fridge and twenty minutes — it adapts a recipe to whatever's actually on hand instead of sending you back to the store.

Be honest about the math

Those numbers don't stack cleanly to 14 hours — they overlap, because planning, shopping, and cooking around what you have are really one connected loop, not three separate chores. A more honest estimate for someone who currently plans meals the old way (manually, Sunday-night-scrambling style) is 2-4 hours back a week, plus noticeably less food going bad in the crisper drawer. That second part matters more than people expect — a chunk of what gets bought never gets used, and most of it is produce nobody got around to before it turned.

There's a quieter money benefit too: a list built from an actual plan tends to run a bit cheaper per trip, because you're not buying "just in case" items or duplicating something already in the pantry because you couldn't remember if you had it.

Set the constraints once, not every week

The part that actually makes this stick is telling the planner your real constraints up front — budget, how many nights you cook versus order in, dietary stuff, how much time you have on a Tuesday versus a Sunday — and letting it hold onto that instead of re-asking every week. Pair it with BudgetTyrant if you want the grocery line item to actually behave itself; that one's built for enforcing spending limits, not just tracking them after the fact.

Where this breaks down

I'll be straight with you: if you genuinely enjoy the ritual of deciding what to cook, or you're feeding one person with simple tastes, this is probably solving a problem you don't have. This is for the household where dinner decisions are a recurring source of friction — multiple people's preferences, a budget that needs watching, and evenings where "let's just order something" keeps winning by default even though it's the expensive option.

The part nobody mentions: leftovers and substitutions

Most meal plans fall apart on Wednesday because something in the plan assumed you'd have an ingredient you don't, or the recipe made too much and nobody wants to eat the same thing twice. RecipeTransformer is specifically there for that failure point — it reworks a recipe around what's actually left in the fridge instead of forcing you back to the original shopping list. That's the detail that makes a plan survive contact with an actual week.

A few things people ask me

Does this work for picky eaters or multiple diets in one house?

Yes — that's actually where it helps most. Tell it who eats what once (vegetarian partner, kid who won't touch anything green, you're doing high-protein) and it builds one plan that works for the table instead of three separate ones.

What if I already meal prep on Sundays?

Then you're already doing the hard part manually. The agent won't replace your Sunday batch-cook — it'll just stop you from standing in the grocery store staring at a blank list trying to remember what you decided three days ago.

Does it actually save money, or just time?

Both, but the money savings is the less obvious one. A list built from an actual plan means fewer impulse buys and less "I'll figure out dinner from whatever's in the cart," which is usually the most expensive way to eat.

Try it for one week

Give an agent your actual constraints — not idealized ones — and let it draft this week's plan and list. If it saves you real time and the fridge has less rotting in it by Sunday, browse the marketplace and let it take over more of the house's weekly admin.