How Appointment Availability Is Calculated

When someone looks for open slots — from the Agenda, the public booking link, or the AI agent — Omnifox doesn't guess: it computes the actual gaps by combining several rules at once.

What the engine combines

1. The calendar's business hours for that day (and its holidays, if any apply).
2. The agent's own shift, for Agent-type calendars — only the overlap of both windows is used.
3. Existing appointments on that calendar (any status except cancelled) — each one blocks its time span plus the service's after-buffer.
4. The service's minimum notice and max advance, which trim the date range that gets offered.
5. Capacity — on a Team calendar or a group service, a slot stays open as long as at least one spot is still free.
6. If the calendar has Google Calendar connected, times already marked busy there are subtracted too.

Slots are offered in fixed blocks (30 minutes by default) inside those free windows.

How double-booking is prevented

The list of free slots is just a suggestion. When the booking is actually confirmed, Omnifox re-checks the real availability of that exact time — with a database lock — before saving the appointment. If two people picked the same slot at nearly the same moment, only the first one gets it; the second sees a message saying the slot was just taken.

Example

A 30-minute service with a 10-minute after-buffer, on a calendar open 9:00–12:00: if there's already an appointment from 10:00 to 10:30, the next real free slot is 10:40 (30-min appointment + 10-min buffer), not 10:30.

Tips

• If the schedule seems to "drift" more than expected between appointments, check the service's buffers.
• The final check always wins — you don't need to refresh the screen to be sure; the system never lets two bookings land on the same slot.

Troubleshooting

See "Troubleshooting Appointments" for the most common error messages (slot just taken, no slots available).