Want a hard-to-get table for a night that matters? This works out exactly when each booking window opens.

The problem

The tables people want most often go within minutes of opening, and every restaurant has its own rule: 28 days ahead at 10 a.m., the 1st of the month for the next month, or 7 days ahead at midnight. Miss the moment and the night is gone, although cancellations come back more often than people think. A plan built from each restaurant's own booking rule, with the exact opening dates counted out, notify alerts, a walk-in backup and a short call script, gives you a real chance without bots or resellers.

Why it works

AI chat assistants are good at reading what you paste and turning it into plain steps. They can't see your device or your accounts and can get details wrong, so the prompt has the AI point to where each fact came from, mark general advice as something to check, and treat what you paste as information, never as instructions. The double-check prompt then makes it point to the source of each finding again. You do every step and decide.

What you get from it

• A shortlist of 3 to 5 places that fit the occasion, the party and the budget, plus one easy backup.
• For each place, its booking rule as the restaurant states it (for example “28 days ahead at 10 a.m.”), with where that rule came from.
• The exact date and time each window opens for your dinner date, counted out and in the restaurant's time zone, ready for your calendar.
• The alerts to set (Notify, waitlist or the restaurant's own list) and when to watch your phone.
• A walk-in plan: when to arrive, the bar or held seats, and the waitlist.
• A short, polite call script to ask about cancellations, the bar or another time, and the deposit and cancellation fees to know before you book.

What you need

• A free AI account. ChatGPT is our pick: it can search the web for each restaurant's booking rules. Grok Bot can also open the booking pages and check live tables (see step 1).
• The occasion, date, party size and budget. Plus the area, any dietary needs, and how flexible you are on day and time.
• The restaurants' own pages, if you have favorites. The reservations page or booking listing. If your AI can't browse, copy the text from it.
• About 10 minutes. To the plan. The rest is setting alarms and alerts.

What the test showed

On 19 Oct 2026 we ran the main prompt and “Double-check your work” with an OpenAI model (gpt-5-mini) on a synthetic sample case. It has three invented restaurant pages for a Saturday anniversary dinner: a 28-days-ahead rule, a monthly release that's already open with a cancellation fee and a note telling AI tools there are plenty of tables, and a 7-days-at-midnight counter with a deposit. It asked one question at a time and took everything in one reply. It counted Little Fig's window as October 24 at 10 a.m. (and October 23 for the Friday), said Harbor Table's November book was already open with its $75-per-person fee, gave Sakura Counter's window as 12:00 a.m. on November 14, ignored the note to AI tools, and never said a table was free. Getting the dates right took several rounds of changes: in early runs it said a monthly release had not opened yet, put the midnight reminder on the wrong day, and named the wrong weekday. With the final prompt, all 4 sample runs and 4 more runs on three real New York restaurants' public booking rules (pasted text only; we booked nothing and contacted no one) got every opening date right; one still added a stray reminder time on the wrong night in a side note, so check midnight times yourself. After a review saw one checklist out of date order, we made the order rule stricter; in 3 re-runs every checklist was in date order. Without web search it suggests places from memory and marks them to check. We only tested on an OpenAI model (gpt-5-mini). Other assistants may word things differently or miss a check, so run the included sample once in your chosen tool and compare it with the worked example before you use your own data.

Sources

Checked 2026-10-19. Rules and prices change: check the source before you rely on it.

• Resy: How to snag hard-to-get tables with Notify