Square Rebooking Gap Finder: Overdue Clients & Drafts
Reads your Square export, finds clients past their own usual visit gap, and drafts a rebooking message for each. Sends nothing.
$9
Works with the AI Agents you already use
Square Rebooking Gap Finder: Overdue Clients & Drafts
Example session with this skill installed
fictional sample data, real script output
$ python3 rebook_gaps.py input-appointments-history.csv --as-of 2026-11-02 --business "Sample Lash Studio" --booking-link https://example.com/book/sample-lash --out output
- Read your context and instructions
- Compiled the square rebooking gap
fictional sample data, real script output (report.txt)
SQUARE REBOOKING GAP FINDER - REPORT
Source file: input-appointments-history.csv
Export type: appointments
As-of date: 2026-11-02
Settings: grace 7 days, one-time after 30 days, lapsed after 365 days, minimum interval 7 days
Columns used: start='start', status='status', service='service', staff='staff', client_name='client_name', phone='client_phone'
Columns not found: email
Rows read: 32 Clients found: 9
Counts: overdue 4, one-time 1, lapsed 1, already booked 1, not due yet 1, too new to judge 1
1. OVERDUE REGULARS (2+ visits, past their own usual gap)
Client Days overdue Last visit Last service Usual gap (days) Gap basis Visits
----------- ------------ ---------- ------------ ---------------- ---------- ------
Fay Sample 89 2026-07-01 Lash Lift 35 1 gap only 2
Ivo Sample 55 2026-08-11 Lash Fill 28 median 3
Ava Sample 27 2026-09-15 Lash Fill 21 median 6
Hana Sample 13 2026-10-06 Brow Tint 14 median 4
2. ONE-TIME CLIENTS (1 visit, no repeat yet)
Client Days since visit Visit date Service
---------- ---------------- ---------- ----------------
Dev Sample 84 2026-08-10 Classic Full Set
3. LAPSED (last visit more than 365 days ago)
Client Days since visit Last visit Last service Usual gap (days) Visits
---------- ---------------- ---------- ------------ ---------------- ------
Gus Sample 447 2025-08-12 Lash Fill 35 3
4. ALREADY BOOKED OR REQUESTED (future appointment or booking request in the file; no message drafted)
Client Next booked Past visits
----------- ----------- -----------
Cleo Sample 2026-11-06 3
5. NOT DUE YET (within their usual gap plus grace)
Client Due around Usual gap (days) Gap basis Visits
---------- ---------- ---------------- --------- ------
Ben Sample 2026-11-17 28 median 4
6. TOO NEW TO JUDGE (1 visit, less than 30 days ago)
Client Visit date
---------- ----------
Eli Sample 2026-10-25
NOTES
- 3 row(s) skipped for status (cancelled, declined, no-show, pending or similar).
- Usual gap = median days between a client's own past visit dates (several rows on one day count as one visit).
- A gap based on 1 gap only is a weak signal. Check it against what you know about the client.
- Nothing was sent. Drafts are in drafts.txt for you to edit and send yourself.
Connects securely to your tools. The creator never sees your data.
About this skill
For salons, barbers, lash and brow studios, nail bars and tattoo shops that book through Square. Download your appointment history from Square, and this skill tells you which regulars are past their usual gap and haven't booked again.
Most of the work is done by a small Python script that comes with the skill. For each client, it works out the usual time between their own past visits (the median gap, so one odd visit doesn't throw it off). Cancelled, declined and no-show appointments are skipped, and two services on one day count as one visit. Anyone with a future booking or booking request already in the file is set aside. The rest are sorted into overdue regulars, ranked by days overdue with their last service, last visit, usual gap and visit count; one-time clients; and long-lapsed clients. Each client in those lists gets a short, plain message draft built from three template files you can edit. The draft uses their first name, last service and your booking link.
You get a plain-text report, a CSV you can sort in a spreadsheet, and a drafts file marked NOT SENT. Your agent explains the results and helps you rewrite the templates in your own voice. You send the messages yourself.
If you only have the Customers export, it still works from First Visit, Last Visit and Transaction Count, but the gaps are rougher (Transaction Count counts payments, not visits) and that file has no future bookings, so check your calendar before sending. The report and drafts say so. Square doesn't publish the columns in these exports, so the script finds them by name, using the names reported by Square community posts and migration guides, and the report says which columns it used and which it couldn't find. It was tested on sample files in that format, not on a real Square download.
The script runs on your computer, has no network code, needs no Square login and changes nothing in Square. Your AI agent reads the report to explain it, so client names in the report pass through whichever AI service runs your agent. It shows who looks due back. It doesn't predict who will rebook, and it doesn't promise bookings or revenue.
Not affiliated with Square or Block, Inc.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Square doesn't publish its export column list, so headers are matched by name and the report lists anything missing. Tested on sample files, not a real Square download. The Customers export gives rough gaps and has no future bookings, so check your calendar before sending. Clients are matched by ID, phone, email, then name. You're responsible for messaging consent and opt-outs.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today
- Software
Python 3.8 or newer
Best export: Square Dashboard, Appointments, Settings, History, Export. Fallback: Customers, Export customers. No accounts, API keys or installs beyond Python.
Permissions
Runs a local Python script that reads the CSV you name and writes three files to an output folder. No network access, no login, no sending.