Skip to main content
Glama

How to connect Squire to ChatGPT, Claude and any AI agent

Let your AI agent run the chair schedule. Cotlet.dev connects your own signed-in Squire account to ChatGPT, Claude and any AI agent over MCP, and to n8n, Make and scripts over REST: appointments, barbers, clients, waitlist and reports for your barbershop.

Unofficial: Cotlet is independent and not affiliated with Squire.

What you can ask

Brief each barber's day at 7 am: every booking with client, services, start and end.

Book waitlist clients into the gaps in tomorrow's schedule, after I approve each booking.

Marcus is out sick today. Move his bookings to another barber with the same time open and the same services.

Block Dev's calendar on Friday from 1 pm to 5 pm for time off.

Show Monday revenue by barber for last week.

Find clients who have not come in for 60 days.

Related MCP server: fresha-mcp

Connect in 2 minutes

  1. Log in to Commander yourself. Make a Cotlet account at cotlet.dev, open Squire in its Cloud Browser, log in to Commander with your own shop login, then press Enable for agents.

  2. Add the Cotlet address. Paste https://cotlet.dev/mcp/v1 into your agent as an MCP server or connector.

  3. Approve the connection. Chat apps send you to a Cotlet sign-in. Code editors, n8n and scripts take an API key from the console.

Squire shares one address, https://cotlet.dev/mcp/v1, with every other app on your Cotlet account. If you enable Squire after connecting the agent, refresh the connector so the Squire tools appear. Free 7-day trial with 2,000 credits, no card needed.

Claude

Settings > Connectors > Add custom connector, URL https://cotlet.dev/mcp/v1/connect.

Claude Code

claude mcp add --transport http cotlet \
  https://cotlet.dev/mcp/v1 \
  --header "Authorization: Bearer $COTLET_API_KEY"

ChatGPT

Settings > Apps & Connectors > Create, MCP server URL https://cotlet.dev/mcp/v1/connect, Authentication: OAuth. No API key is needed.

Cursor

Copy .cursor/mcp.json from this repository into your project or ~/.cursor/mcp.json:

{
  "mcpServers": {
    "cotlet": {
      "url": "https://cotlet.dev/mcp/v1",
      "headers": { "Authorization": "Bearer ${env:COTLET_API_KEY}" }
    }
  }
}

Codex

# ~/.codex/config.toml
[mcp_servers.cotlet]
url = "https://cotlet.dev/mcp/v1"
bearer_token_env_var = "COTLET_API_KEY"

n8n, Make and Zapier

  • n8n: MCP Client Tool inside an AI Agent, Transport HTTP Streamable, URL https://cotlet.dev/mcp/v1, Auth Bearer with your Cotlet key, tool squire_read.

  • Make: MCP Client > Call a tool. Zapier: MCP Client by Zapier > Run Tool. Server URL https://cotlet.dev/mcp/v1, your Cotlet key as the token, tool squire_write.

Squire REST API

Create a Cotlet API key with access to Squire and store it as COTLET_API_KEY in your environment or your automation platform's credential store, outside shared workflow exports.

curl https://cotlet.dev/api/v1/squire/barbers \
  -H "Authorization: Bearer $COTLET_API_KEY"

A successful call puts the tool output in result. MCP and REST carry the same 11 tools, inputs and answers, with one API key and one price. Agents use MCP; scripts, sheets and HTTP steps in n8n, Make or Zapier use REST at https://cotlet.dev/api/v1/squire.

Squire tools through Cotlet.dev

6 read, 5 write. Over MCP each tool is an operation of squire_read or squire_write.

Tool

What it does

Access

REST endpoint

list_barbers

List the barbers on your shop's team

read

GET /api/v1/squire/barbers

list_services

List the services on your shop's service menu

read

GET /api/v1/squire/services

list_appointments

List appointments for a day or a date range, optionally for one barber

read

GET /api/v1/squire/appointments

create_appointment

Book an appointment for a client with a barber

write, asks first

POST /api/v1/squire/clients/{client_id}/appointments

reschedule_appointment

Move an appointment to a new time or another barber

write, asks first

POST /api/v1/squire/appointments/{appointment_id}/reschedule

cancel_appointment

Cancel a client's appointment

write, asks first

POST /api/v1/squire/appointments/{appointment_id}/cancel

block_time

Block a barber's calendar for a time window

write

POST /api/v1/squire/barbers/{barber_id}/time/block

remove_block

Remove a blocked time window from a barber's calendar

write

DELETE /api/v1/squire/blocks/{block_id}

list_waitlist

List clients on the shop's waitlist

read

GET /api/v1/squire/waitlists

list_clients

List or search your shop's clients

read

GET /api/v1/squire/clients

get_revenue_report

Read revenue and client retention from the Reporting Dashboard for a date range

read

GET /api/v1/squire/revenue-report

Times are in the shop's time zone. A slot that is booked or blocked fails with SLOT_TAKEN and nothing changes. Squire lists the Wait List under its PRO plan; on other plans the list comes back empty. Phone and email come back only when you ask for them. Revenue is in the shop's currency.

Approvals and undo

Nothing reaches a client before you approve. 3 of the 5 write tools ask first: create_appointment, reschedule_appointment and cancel_appointment wait in your Cotlet console. block_time and remove_block run at once, because a block reaches no client and undoes in one call. Reads run at once too. For a scheduled run you can switch the wait off per tool.

Action

Undo

Book an appointment

Cancel it; Squire may already have sent the confirmation

Reschedule an appointment

Move it back; the client may hear about both changes

Cancel an appointment

Book it again; a no-show or late fee is never charged by a tool

Block time

Remove the block

Remove a block

Block the window again

A block does not clear the bookings already in its window. Ask the agent to list them first. When a whole day moves, let the agent go one booking at a time, so each client hears about a single change.

Run it locally (stdio)

This repository is a small MCP server for clients that only start local processes. It speaks MCP over stdio, lists the 11 Squire tools without any key, and forwards every tool call to the hosted Cotlet.dev server as squire_read or squire_write. The hosted URL above stays the recommended setup. Python 3.10+, no dependencies.

# COTLET_API_KEY holds your Cotlet API key
claude mcp add squire -e COTLET_API_KEY="$COTLET_API_KEY" -- \
  uvx --from git+https://github.com/cotlet-connect/squire-mcp squire-mcp

mcp.json in this repository has the same server for clients that read a JSON config. Or with Docker:

docker build -t squire-mcp .
docker run -i --rm -e COTLET_API_KEY squire-mcp

Without COTLET_API_KEY a tool call answers with where to create a key.

Command line

python3 squire_agent.py connect                     # hosted setup for Claude Code, Cursor, Codex and chat apps
python3 squire_agent.py tools                       # the 11 Squire tools
python3 squire_agent.py call list_barbers           # run one tool through Cotlet.dev
python3 squire_agent.py call list_appointments '{"date_from": "2026-11-16"}'
python3 squire_agent.py serve                       # stdio MCP server (same as no argument)

Does Squire have an API or MCP server?

No. Squire has no MCP server, no public API and no developer docs (checked 28 Oct 2026). Zapier, n8n and Pipedream have no Squire app, and clients can be exported by hand from Commander only as a snapshot. On GitHub, repos named squire-mcp are coding tools of the same name, not servers for Squire barbershops. Cotlet runs 11 Squire tools hosted, over MCP for agents and REST for scripts.

Privacy and access

  • What Cotlet stores: your logged-in Commander profile, AES-256-GCM encrypted, in Finland. Never the password.

  • What your agent gets: 11 Squire tools. No cookie, no login.

  • What the session sees: what your Commander login sees: the whole shop for an owner, one chair for a barber.

  • What no tool touches: checkout, cards on file, tips, deposits, booth rent and payouts.

  • If Squire asks for a code: you type it in yourself.

  • How to cut access: disable for agents, reset the browser or delete the key. The next call fails.

Client phone numbers and emails sit in the same Commander login, so give the agent a Squire user that sees only what it needs. Read Squire's terms in your own browser and look for rules on bots, automated access and shared logins before you connect.

Price

10 credits per successful call, MCP or REST, and 50 credits a day for the Cloud Browser. One price for every barber in the shop. Only successful calls use credits; failed calls and timeouts cost nothing. Credit packs start at 5,000 credits for $29. Your Squire plan is billed by Squire as before. Full pricing

Questions from Squire users

How do I connect Squire to ChatGPT or Claude?

Sign up at cotlet.dev, log in to Squire once in its Cloud Browser and add Cotlet to ChatGPT or Claude, following the steps in your console. No API key is needed. ChatGPT and Claude then work in your Squire account through Cotlet, in their phone apps too.

Which Squire plan do I need?

Cotlet works in your own Squire account, so it can only do what your plan shows in Commander. Squire's pricing page lists the waitlist and multiple barber accounts under its PRO plan and the Reporting Dashboard from the first plan (checked 28 Oct 2026).

Can it take payments or issue refunds?

No. Payments, payouts, rent collection and card processing stay in Squire. The tools book, move and cancel appointments and read reports.

Does it work for an independent barber?

Yes. Squire has accounts for independent barbers, and the tools then see your own chair: your bookings, blocks, clients and revenue.

Can I use the Squire API without an AI agent?

Yes. Call Cotlet's REST API from a script or an HTTP step in n8n, Make or Zapier with your Cotlet key.

Find it on

Cotlet.dev is run by Take Group sp. z o.o., Katowice, Poland. Questions about the Squire tools: contact@cotlet.dev. Squire is a trademark of its owner.

More: https://cotlet.dev/integrations/squire-api-mcp-server?utm_source=github&utm_medium=readme&utm_campaign=squire

Start free: https://cotlet.dev/app/signup?app=squire&utm_source=github&utm_medium=readme&utm_campaign=squire

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    • Squire MCP server for ChatGPT, Claude and any AI agent: list appointments, book appointment, reschedule appointment, cancel appointment, list waitlist, block time off, list clients and get revenue report in your own Squire account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to Squire yourself; Cotlet does not store your password and you need no Squire API key. About the official Squire API: Squire has no public API and no developer docs (checked 28 Oct 2026). The only mention of an API is in its privacy policy, about third-party sites such as Google and Instagram that connect to Squire. Setup and FAQ: https://cotlet.dev/integrations/squire-api-mcp-server

      https://cotlet.dev/mcp/v1/squire
    • Acuity Scheduling MCP server for ChatGPT, Claude and any AI agent: list appointments, check availability, book appointment, reschedule appointment, cancel appointment and block off time in your own Acuity Scheduling account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to Acuity Scheduling yourself; Cotlet does not store your password and you need no Acuity Scheduling API key. About the official Acuity Scheduling API: Acuity Scheduling has a full REST API but no official MCP server (checked 17 Oct 2026). The API uses HTTP Basic auth with your user ID and API key, and Acuity says API access is available on the Premium plan (formerly Powerhouse), so Starter and Standard accounts have no key to give an agent. Setup and FAQ: https://cotlet.dev/integrations/acuity-scheduling-api-mcp-server

    • Fresha MCP server for ChatGPT, Claude and any AI agent: list appointments, create appointment, reschedule appointment, list clients and get sales summary in your own Fresha account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to Fresha yourself; Cotlet does not store your password and you need no Fresha API key. About the official Fresha API: Fresha has no public API and no developer docs (checked 11 Oct 2026). Its only official data connection is the paid Data Connector add-on, which syncs your data out to reporting tools through a Snowflake database connection. Setup and FAQ: https://cotlet.dev/integrations/fresha-api-mcp-server

      https://cotlet.dev/mcp/v1/fresha
    • DaySmart MCP server for ChatGPT, Claude and any AI agent: list appointments, book appointment, find client, update pet record and export sales in your own DaySmart account. Cotlet works in the account through your Cloud Browser, a real Chrome where you sign in to DaySmart yourself; Cotlet does not store your password and you need no DaySmart API key. About the official DaySmart API: DaySmart Salon and DaySmart Pet have no public API (checked 11 Oct 2026). DaySmart advertises scheduling API access only for its separate Appointments product. Setup and FAQ: https://cotlet.dev/integrations/daysmart-api-mcp-server

    Related MCP Servers

    • A
      license
      A
      quality
      C
      maintenance
      Connects a signed-in Appointy account to AI agents so users can book, reschedule, cancel appointments, send reminders, and inspect staff schedules using natural language.
      7
      Elastic 2.0
    • A
      license
      A
      quality
      C
      maintenance
      Enables AI agents to manage appointments, reschedule bookings, look up clients, and add client notes in a user's signed-in business account without requiring an API key.
      7
      Elastic 2.0
    • A
      license
      A
      quality
      C
      maintenance
      Enables AI agents to work in a user's signed-in GlossGenius account, allowing natural-language booking, rescheduling, schedule queries, and client follow-ups without requiring a GlossGenius API key.
      7
      Elastic 2.0
    • A
      license
      A
      quality
      B
      maintenance
      Connects AestheticsPro accounts to ChatGPT, Claude, and other AI agents through Cotlet, enabling appointment scheduling, client contact access, lead creation, and staff assignment with user approval.
      7
      Elastic 2.0