Cleaning businesses already use software to manage bookings, customers, schedules, services, payments, and reports. Until now, working with that information usually meant opening the right dashboard, finding the right record, applying filters, and making changes manually.
Cleanly MCP adds another way to work.
Cleanly can now connect with compatible AI assistants such as ChatGPT and Claude, allowing authorized users to ask questions and perform supported business tasks through natural conversation.
You can check this week’s schedule, find available time slots, quote a cleaning job, review customer balances, compare revenue, reschedule a booking, create a discount code, or inspect business performance without first navigating through multiple screens.
The AI assistant provides the conversational interface. Cleanly remains the system that controls the bookings, business data, permissions, and underlying operations.
What Is Cleanly MCP?
Cleanly MCP is a secure remote Model Context Protocol server built directly into Cleanly Booking.
Model Context Protocol, commonly called MCP, gives compatible AI applications a structured way to discover and use tools from another application. Instead of copying Cleanly data into an AI conversation manually, an authorized assistant can request supported information directly from Cleanly.
The current implementation includes 39 core MCP tools covering bookings, customers, schedules, quotes, availability, reports, services, coupons, branches, job status, and other business operations. Additional tools become available for refunds when that capability is enabled and for team management when the Cleanly Membership add-on is installed.
Users do not need to know those tool names.
A compatible AI client can determine which Cleanly tool is appropriate from a normal request such as:
What is on the schedule this week?
Quote a three-bedroom deep clean with oven cleaning.
Which customers still owe us money?
Move Tuesday morning’s booking to Thursday afternoon.
This creates a Cleanly Conversational Operations Layer between the user and the booking system. You communicate through natural language while Cleanly continues to handle the underlying business logic.
Why Does Cleanly MCP Matter for a Cleaning Business?
Cleanly MCP lets an AI assistant work with authorized business information instead of relying only on whatever a user manually types into the conversation.
A general AI assistant does not automatically know tomorrow’s bookings, current service prices, available appointment slots, customer balances, or this month’s revenue.
Once Cleanly is securely connected, those questions can be answered using current Cleanly data.
The same principle applies to supported actions. An authorized user can ask the assistant to reschedule a job, add a booking note, create a coupon, or perform another permitted task.
The important distinction is control.
Connecting Cleanly does not give an AI assistant unrestricted access to WordPress. The administrator approves specific permissions, and the assistant can only use the Cleanly capabilities available within those permissions.
What Can You Do With Cleanly MCP?
Cleanly MCP covers many of the daily operations that cleaning businesses already manage inside Cleanly.
The available capabilities depend on the permissions granted to the connected AI client and the Cleanly features installed on the website.
| Business area | What you can do through Cleanly MCP |
| Bookings | Search, view, create, edit, reschedule, and add notes |
| Availability | Check genuinely available dates and time slots |
| Quotes | Generate detailed booking quotes without saving a booking |
| Customers | Review history, lifetime value, balances, and upcoming jobs |
| Schedule | View the cleaning diary for a day or date range |
| Reports | Review revenue, financials, trends, service performance, and occupancy |
| Services | Create and update supported catalog items |
| Coupons | Validate, create, and update discount codes |
| Branches | View, create, and update business branches |
| Jobs | Mark jobs finished or reopen them |
| Payments | Manage supported payment states and controlled refunds |
| Team | Review and assign staff when Membership is installed |
The result is more than an AI chatbot answering generic questions. The assistant can work with the operational system behind the business when the user has authorized it to do so.
Can AI Create and Manage Cleaning Bookings?
Authorized AI assistants can create and manage bookings through Cleanly’s existing booking system.
This is an important part of how Cleanly MCP is designed.
AI-created bookings do not go through a separate booking engine. Cleanly routes the request through its own booking code, keeping the booking inside the same system used for normal Cleanly bookings.
A conversation could begin with:
Quote a deep clean for a three-bedroom home with oven cleaning.
Cleanly can calculate the quote without saving a booking or reserving the requested slot.
If the customer wants to proceed, the next request could be:
Book it for Sarah Chen for the 24th and do not email her yet.
Existing bookings can also be searched, updated, rescheduled, and supplemented with notes when the connected client has the required permissions.
Can AI Check Availability and Quote Cleaning Jobs?
Cleanly MCP checks availability and pricing against Cleanly’s configured booking rules rather than asking the AI assistant to estimate them.
Availability can depend on time slots, capacity, closed days, lead times, cut-off rules, and other booking conditions. Cleanly evaluates those rules when an assistant asks whether a date or time is available.
That makes questions such as these practical:
Is there availability for a deep clean on the 24th?
Where does next Friday still have room?
The quoting workflow works in a similar way. Cleanly uses the configured service catalog, pricing, extras, and applicable discounts to return a full price breakdown.
Compatible clients that support MCP Apps can also display structured quote information directly inside the conversation.
Can AI Help With Customer Management?
Cleanly MCP gives authorized users conversational access to customer history and relevant booking information.
Instead of searching through records manually, a user can ask about a specific customer or search across customer information.
Cleanly can return details such as previous jobs, upcoming bookings, lifetime value, total job count, and outstanding balance.
For example:
What does the Patel household have booked next?
What is this customer’s lifetime value?
Which customers still have an outstanding balance?
The answer comes from Cleanly data rather than from a customer list copied manually into the AI conversation.
Can AI Analyze Cleaning Business Reports?
Cleanly MCP makes financial and operational reports available through natural-language questions.
The reporting tools cover dashboard summaries, revenue, financial reports, booking trends, service performance, and occupancy.
Users can ask questions such as:
Show my dashboard summary for this month and compare it with last month.
Which cleaning extras are selling the most?
How has booking volume changed over the last 30 days?
Where do we still have unused capacity?
Cleanly supports common reporting ranges as well as validated custom date ranges. Reporting can also distinguish first-time and returning customers and show how booked jobs compare with available capacity.
In clients that support Cleanly’s interactive MCP Apps, some report results can appear as structured metrics, charts, and category breakdowns inside the conversation.
This is where conversational access becomes especially useful. Instead of deciding which report to open first, the business owner can start with the question they actually want answered.
Can Cleanly MCP Manage Services and Discount Codes?
Authorized AI clients can create and update supported service catalog items and discount codes.
Cleanly MCP can work with cleaning types, frequencies, home attributes, extras, prices, discounts, and service durations when the connected client has catalog permissions.
A business owner could request:
Increase the deep clean price to 180.
Or:
Create a 15% discount code for October with one use per customer.
Deletion is handled more carefully. Cleanly checks whether a catalog item is referenced by existing bookings before allowing it to be removed, and eligible deletions require a preview before confirmation.
Can AI Help With Cleaner Scheduling?
Team scheduling tools are available through Cleanly MCP when the Cleanly Membership add-on is installed.
Membership-enabled sites can let authorized AI clients list cleaners, review staff reports, and assign team members to bookings.
A manager might ask:
Show the team’s workload for the last 30 days.
Or:
Assign Maria to Thursday’s deep clean.
Cleanly can reject an assignment that would double-book a cleaner when that conflict is not allowed.
When Membership is not installed, those team tools are simply not published to the AI client. The available tool catalog therefore reflects what the website can actually do.
How Does Cleanly MCP Protect Business Data?
Cleanly MCP uses scoped OAuth authorization to limit what each connected AI client can access and change.
Only a WordPress user with the required Cleanly MCP capability can authorize a connection. Cleanly uses an OAuth authorization flow with PKCE protection, scoped access tokens, rotating refresh tokens, token revocation, and WordPress capability checks.
Permissions are separated by purpose.
A client with read access does not automatically receive permission to edit bookings, delete records, change booking settings, or issue refunds.
Sensitive credentials are also intentionally excluded. Published MCP tools do not expose payment gateway keys, SMS credentials, mailbox credentials, API keys, licence information, or Cleanly’s MCP and OAuth security settings.
The goal is delegated access to specific Cleanly operations, not unrestricted access to the WordPress website.
How Does Cleanly Protect Refunds and Deletion?
Refunds and destructive actions require stronger controls than ordinary Cleanly requests.
Refund functionality is disabled by default. When it is disabled, refund tools are not offered to connected AI assistants at all.
If an administrator enables refunds, the assistant still cannot immediately move money. The refund must first be previewed so the user can see what will happen before explicitly confirming the action.
Confirmation tokens are short-lived and tied to the exact user, action, record, and current revision. If the record changes after the preview, the operation must be reviewed again.
Cleanly also waits for the original card payment provider to confirm that money was returned before marking the booking as refunded.
Booking deletion follows a similar preview-first process. Records that form part of protected financial history cannot be deleted.
These controls allow conversational workflows without treating high-impact actions like ordinary requests.
How Do You Connect Cleanly to ChatGPT?
Cleanly connects to supported ChatGPT environments through the MCP server URL generated inside the AI Connect dashboard.
Start in Cleanly → Settings → AI Connect. Enable the AI endpoint and automatic app registration, then save the settings.
Next, open Cleanly → Dashboard → AI Connect and copy the displayed MCP server URL. Using the generated address is important because Cleanly accounts for the site’s WordPress configuration, including subdirectories, multisite paths, and REST settings.
In a ChatGPT environment that supports custom MCP apps, create a custom app with the Cleanly server URL and OAuth authentication. WordPress then opens an authorization screen where the administrator can review the requested permissions before approving access.
A simple first test could be:
Use Cleanly to show my dashboard summary for this month.
Availability of custom MCP apps and write actions can depend on the user’s ChatGPT plan and workspace policy.
How Do You Connect Cleanly to Claude?
Cleanly supports compatible Claude environments through remote custom connectors.
Copy the MCP server URL from Cleanly → Dashboard → AI Connect and add it as a custom connector in Claude or Claude Desktop.
The connection opens the WordPress authorization process, where the administrator reviews and approves the requested Cleanly permissions.
After connecting, a simple test could be:
Use Cleanly to list my branches.
Remote connector availability can depend on the user’s Claude plan and organization settings.
What Does Cleanly MCP Intentionally Not Allow?
Cleanly MCP excludes several high-risk or unsupported operations by design.
The current implementation does not support bulk destructive deletion, partial refunds, refunds for non-card gateways, retrieving payment or API credentials, charging a customer’s card, or unrestricted editing of MCP and OAuth security controls.
An assistant also cannot simply change a booking status to refunded without the original payment provider confirming that the refund actually happened.
Protected financial-record bookings cannot be deleted.
These boundaries are deliberate. Giving AI access to business software should make routine operations easier without removing important controls around money, credentials, and permanent changes.
A New Way to Work With Cleanly
Cleanly MCP introduces conversational business operations without replacing the booking system behind them.
Bookings remain Cleanly bookings. Quotes use Cleanly pricing. Availability follows Cleanly rules. Reports come from Cleanly business data. Administrators remain in control of the permissions granted to each connected AI client.
What changes is the interface.
Instead of always beginning with a dashboard, menu, filter, or record search, users can begin with a question or instruction.
That is the idea behind the Cleanly Conversational Operations Layer.
AI can move beyond giving general cleaning business advice and, with permission, work directly with the operational workflows already managed inside Cleanly.
For cleaning businesses exploring practical AI in 2026, this creates a more useful relationship between AI assistants and the software that runs day-to-day operations.
Frequently Asked Questions About Cleanly MCP
Cleanly MCP is the remote Model Context Protocol integration built into Cleanly Booking. It allows authorized compatible AI clients to work with supported Cleanly bookings, schedules, customers, reports, services, branches, and other business operations.
Cleanly MCP can help users check schedules and availability, create quotes, create or reschedule bookings, review customers, analyze reports, manage services and discount codes, update branches, and perform other supported actions according to the permissions granted.
Yes. Cleanly can connect to ChatGPT environments that support custom remote MCP apps. The connection uses the MCP server URL generated in Cleanly’s AI Connect dashboard and an OAuth authorization process.
Yes. Cleanly supports compatible Claude environments through remote custom connectors using the Cleanly MCP server URL and WordPress OAuth authorization.
Cleanly MCP uses scoped OAuth permissions, PKCE protection, WordPress capability checks, token controls, sensitive credential exclusions, revision-safe updates, and additional confirmation requirements for high-impact actions.
No. Users can communicate with a compatible AI assistant using natural language. The client can choose an appropriate Cleanly tool from the tools and schemas available to it.
