Local services & SaaS
CleanManager
Cleaning tasks, check-ins, customers, cleaning plans, audits, storage and wages in CleanManager.
29 endpoints
Sweepivo publishes no public API and no MCP server (checked 15 Nov 2026). Cotlet runs a hosted Sweepivo MCP server with 10 tools that ChatGPT and Claude use in your own Sweepivo account. Cleaner assignments, price changes and website pages wait for your OK.
10 tools Read and write
Sweepivo publishes no public API, no developer documentation and no API keys. No Sweepivo MCP server and no official ChatGPT or Claude connector was found.
| What you need | Sweepivo API | By hand | Cotlet |
|---|---|---|---|
| Ask ChatGPT or Claude to do the work | No API or MCP server | Copy and paste | Hosted MCP server |
| Bookings, assignments, payments, payroll | Not available | One screen at a time | 10 tools, writes after your OK |
| Run every morning on a schedule | Not available | Only when you open the app | REST or n8n, same key |
Checked 15 Nov 2026. Sources: Sweepivo home and features · Sweepivo llms.txt · Sweepivo pricing · Sweepivo app tour · Sweepivo sign-in
Once, with your emailed code, in a browser we host.
One address in its settings.
Open visits, new bookings, unpaid jobs.
Your agent works in your own signed-in Cloud Browser. Anything that reaches your cleaners, your prices or your public website waits for your OK. Who can see your session
Sign in to Sweepivo in a Cloud Browser. Open sweepivo.com/signin in your Cotlet Cloud Browser and sign in as the owner or a manager, with the emailed code or your passkey.
Add the Cotlet address. Enter https://cotlet.dev/mcp/v1 as the MCP server or connector URL in your agent.
Let the agent in. Chat apps open a Cotlet sign-in page. Editors, n8n and scripts send an API key made in the console.
Sweepivo uses the shared address https://cotlet.dev/mcp/v1 with every other app you turn on, so one connector covers them all. Added Sweepivo later? Refresh the connector and the 10 Sweepivo tools show up.
Settings > Apps & Connectors > Create
MCP server URL: https://cotlet.dev/mcp/v1/connect
Authentication: OAuth 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" // ~/.cursor/mcp.json
{
"mcpServers": {
"cotlet": {
"url": "https://cotlet.dev/mcp/v1",
"headers": { "Authorization": "Bearer ${env:COTLET_API_KEY}" }
}
}
} # ~/.codex/config.toml
[mcp_servers.cotlet]
url = "https://cotlet.dev/mcp/v1"
bearer_token_env_var = "COTLET_API_KEY" MCP Client Tool (inside an AI Agent)
Transport: HTTP Streamable
URL: https://cotlet.dev/mcp/v1
Auth: Bearer, <your key>
Tool: sweepivo_read One tool per part of Sweepivo that an owner works in. Reads run at once; the three writes ask first.
The owner home with uncovered jobs, bookings to approve and unpaid visits, then the cleaner on each visit.
Returns what Sweepivo's owner home shows for the day: the jobs, which are done, what has been collected, and the items that need attention, which Sweepivo lists as uncovered jobs, bookings to approve and unpaid visits. It changes nothing and runs without asking.
MCP callsweepivo_read → view_daily_tasks
REST endpointGET https://cotlet.dev/api/v1/sweepivo/daily-tasks/view
| Name | Type | Description |
|---|---|---|
| date string | string | ISO date. Defaults to today. |
{
"date": "2026-11-16",
"visits": 9,
"done": 0,
"collected": 0,
"needs_you": {
"uncovered_jobs": 2,
"bookings_to_approve": 1,
"unpaid_visits": 3
}
} Sweepivo checks the schedule for conflicts and sends the cleaners a reassignment notice when a visit moves, so this reaches your team. The agent names the visit and the cleaners before it saves. By default you confirm each call; you can switch that off for this tool.
MCP callsweepivo_write → assign_cleaners
REST endpointPOST https://cotlet.dev/api/v1/sweepivo/cleaners/assign
| Name | Type | Description |
|---|---|---|
| visit string required | string | Visit or booking ID, from view_daily_tasks or view_cleaning_bookings. |
| cleaners array required | array | One or more cleaners or a team, by the names Sweepivo shows. Several names make a multi-cleaner job. |
{
"visit": "bk_2041",
"cleaners": [
"Cleaner A"
]
} {
"visit": "bk_2041",
"service": "Pressure washing",
"start": "2026-11-18T09:00",
"cleaners": [
"Cleaner A"
],
"conflicts": 0,
"saved": true
} What clients booked through your link, the price Sweepivo quoted and the answers behind it.
Sweepivo quotes each online booking instantly and lets the client pick a real open time. If you review requests before they reach the calendar, `to_approve` lists the ones waiting for you. Read only.
MCP callsweepivo_read → view_cleaning_bookings
REST endpointGET https://cotlet.dev/api/v1/sweepivo/cleaning-bookings/view
| Name | Type | Description |
|---|---|---|
| from string | string | ISO date to start from. Defaults to today. |
| to string | string | ISO date to end at. Defaults to seven days after `from`. |
| status string | string | One of `all`, `to_approve`, `confirmed`. Defaults to `all`. |
{
"bookings": [
{
"id": "bk_2041",
"client": "Example Client A",
"service": "Pressure washing",
"price": 289,
"start": "2026-11-18T09:00",
"status": "to_approve"
},
{
"id": "bk_2042",
"client": "Example Client B",
"service": "Window cleaning",
"price": 180,
"start": "2026-11-19T13:00",
"status": "confirmed"
}
]
} Your booking questions are the ones you would ask on the phone: how many windows, inside too, how big the pool, is there a spa. In Sweepivo each answer adds price and time, so this tool returns every answer with what it added. Read only.
MCP callsweepivo_read → view_booking_answers
REST endpointGET https://cotlet.dev/api/v1/sweepivo/booking-answers/view
| Name | Type | Description |
|---|---|---|
| booking string required | string | Booking ID from view_cleaning_bookings. |
{
"booking": "booking"
} {
"booking": "bk_2042",
"service": "Window cleaning",
"answers": [
{
"question": "How many windows?",
"answer": "18",
"added_price": 144,
"added_minutes": 90
},
{
"question": "Inside too?",
"answer": "No",
"added_price": 0,
"added_minutes": 0
},
{
"question": "Screens and tracks?",
"answer": "Yes",
"added_price": 36,
"added_minutes": 30
}
],
"price": 180
} Read only. The agent reads paid and unpaid visits and pay per job; it charges no card and starts no pay run.
Card payments in Sweepivo go through your own Square account, and Sweepivo keeps the paid and unpaid state per visit. This tool reads that state; it does not charge a card or send a reminder.
MCP callsweepivo_read → track_customer_payments
REST endpointGET https://cotlet.dev/api/v1/sweepivo/customer-payments/track
| Name | Type | Description |
|---|---|---|
| status string | string | One of `unpaid`, `paid`, `all`. Defaults to `unpaid`. |
| from string | string | ISO date of the earliest visit to include. |
{
"status": "unpaid",
"visits": [
{
"visit": "bk_1988",
"client": "Example Client C",
"service": "Deep clean",
"date": "2026-11-12",
"amount": 240
},
{
"visit": "bk_1994",
"client": "Example Client D",
"service": "Move-out clean",
"date": "2026-11-13",
"amount": 320
}
],
"total_unpaid": 560
} Sweepivo pays hourly, a fixed amount or a percentage per job, reconciled per job. The tool returns each cleaner's jobs and pay so you can check them before a pay run. It does not start the pay run. Payroll comes with Sweepivo's Shine plan and above.
MCP callsweepivo_read → review_cleaner_payroll
REST endpointGET https://cotlet.dev/api/v1/sweepivo/cleaner-payrolls/review
| Name | Type | Description |
|---|---|---|
| from string required | string | ISO date, first day of the pay period. |
| to string required | string | ISO date, last day of the pay period. |
| cleaner string | string | One cleaner. Leave empty for the whole team. |
{
"from": "from",
"to": "to"
} {
"from": "2026-11-09",
"to": "2026-11-15",
"cleaners": [
{
"cleaner": "Cleaner A",
"jobs": 11,
"pay_type": "hourly",
"pay": 682
},
{
"cleaner": "Cleaner B",
"jobs": 8,
"pay_type": "percentage",
"pay": 512
}
]
} Returns the figures Sweepivo's reports show at a glance: revenue, labour cost and the services booked. Read only.
MCP callsweepivo_read → view_business_reports
REST endpointGET https://cotlet.dev/api/v1/sweepivo/business-reports/view
| Name | Type | Description |
|---|---|---|
| from string required | string | ISO date. |
| to string required | string | ISO date. |
{
"from": "from",
"to": "to"
} {
"from": "2026-10-01",
"to": "2026-10-31",
"revenue": 14820,
"labour_cost": 6930,
"services_booked": [
{
"service": "Standard clean",
"count": 64
},
{
"service": "Deep clean",
"count": 12
},
{
"service": "Window cleaning",
"count": 9
}
]
} What new clients see and pay. Every change waits for your OK.
Sweepivo prices by bedroom and bathroom count, square feet, window, car space, story or room, with extras charged once or per item, frequency discounts and minimums. New prices reach every client who books after the change, so the agent shows you the old and new value first. By default you confirm each call; you can switch that off for this tool.
MCP callsweepivo_write → set_service_prices
REST endpointPUT https://cotlet.dev/api/v1/sweepivo/service-prices
| Name | Type | Description |
|---|---|---|
| action string required | string | One of `list`, `update`. `list` only reads. |
| service string | string | Service name, for `update`, for example Standard clean. |
| changes object | object | For `update`: the price, extra, minimum charge or frequency discount to change, in Sweepivo's own labels. |
{
"action": "action"
} {
"service": "Standard clean",
"changed": {
"extra": "Inside the oven",
"from": 30,
"to": 35
},
"saved": true
} Sweepivo builds a branded site with a page for every service and town you serve. `update` saves a draft; `publish` puts it on the public site, which is why the agent shows you the page first. By default you confirm each call; you can switch that off for this tool.
MCP callsweepivo_write → create_cleaning_website
REST endpointPOST https://cotlet.dev/api/v1/sweepivo/cleaning-websites
| Name | Type | Description |
|---|---|---|
| action string required | string | One of `list`, `update`, `publish`. `list` only reads. |
| page string | string | A service page or town page, for `update`. |
| text string | string | For `update`: the words for the page, yours or drafted by Sweepivo's AI writer. |
{
"action": "action"
} {
"page": "Deep cleaning in Example Town",
"action": "update",
"published": false,
"saved": true
} Your signed-in Sweepivo browser profile, AES-256-GCM encrypted, in Finland. Sweepivo has no password to store: you sign in with an emailed code or a passkey.
The 10 Sweepivo tools. The session cookie stays in the Cloud Browser.
The visits, bookings, payments and payroll your Sweepivo owner or manager login can open.
The lockbox and door-code vault, your Square connection, two-step sign-in settings, Hiring Studio and your Sweepivo plan.
You sign in again yourself with a fresh code, in the same Cloud Browser.
Disable Sweepivo for agents in the console, or reset the Cloud Browser. Either ends the session.
Cotlet never sees or stores your password. You type the Sweepivo code yourself; the agent only uses the session that sign-in leaves behind. Our site scores A+ on Qualys SSL Labs, securityheaders.com and Mozilla Observatory (live results).
Cotlet watches the apps it serves and updates the Sweepivo server when Sweepivo changes its screens, so your agent keeps working without a change on your side.
Your plan decides what the agent can see. Payroll arrives with Shine; Sweepivo AI, route planning and Hiring Studio with Sparkle. A tool returns nothing for a part of Sweepivo your plan does not include.
If Sweepivo signs you out, one click in the Cotlet console reopens the Cloud Browser so you can sign in again.
| Action | How to put it right |
|---|---|
| Wrong cleaner on a visit | Call assign_cleaners again with the right one; Sweepivo sends a reassignment notice |
| Price or extra changed | Call set_service_prices again with the old value |
| Website page saved | Update it again; a page that was never published was seen by nobody |
| Website page published | Already public: update and publish the corrected text |
All 3 write tools ask first: by default you confirm each call, and you can switch that off per tool. The 7 read tools run at once, and none of them charges a card, sends a reminder or starts a pay run.
{
"name": "assign_cleaners",
"arguments": {
"visit": "bk_2041",
"cleaners": [
"Cleaner A"
]
}
} curl -X POST https://cotlet.dev/api/v1/sweepivo/cleaners/assign \
-H "Authorization: Bearer $COTLET_API_KEY" \
-H "Content-Type: application/json" \
-d '{"visit":"bk_2041","cleaners":["Cleaner A"]}' {
"visit": "bk_2041",
"service": "Pressure washing",
"start": "2026-11-18T09:00",
"cleaners": [
"Cleaner A"
],
"conflicts": 0,
"saved": true
} The example gives the $289 pressure washing booking to Cleaner A after Sweepivo found no conflicts. A cron job or n8n flow can make the same call over REST at /api/v1/sweepivo with the same key and price, in JSON. Failed calls are free.
About $19 a monthfor a cleaning company with a small crew, 3,230 credits. Your Sweepivo plan stays as it is.
| Job | Calls | Credits |
|---|---|---|
| Morning look at the owner home | 1 × 26 | 260 |
| New bookings and their answers | 3 × 26 | 780 |
| Cover open visits | 2 × 26 | 520 |
| Unpaid visits list | 2 × 4 | 80 |
| Payroll check before the pay run | 1 × 4 | 40 |
| Monthly revenue and labour report | 1 × 1 | 10 |
| Price changes and the booking link | 2 × 2 | 40 |
| Cloud Browser, 30 days | 1,500 | |
| Total | 173 calls | 3,230 |
The smallest pack, 5,000 credits for $29, covers this month with credits left over. More cleaners and more bookings mean more calls.
Credit packs are one-time purchases, not a subscription. The month above is an example built from our own assumptions, not measured customer usage.
| Credits | Price, one time | Calls at 10 credits | Per 1,000 credits |
|---|---|---|---|
| 5,000 | $29 | 500 | $5.80 |
| 15,000 | $79 | 1,500 | $5.27 |
| 32,000 | $149 | 3,200 | $4.66 |
| 60,000 | $249 | 6,000 | $4.15 |
Only successful calls use credits; failed calls and timeouts cost nothing. At a balance of 0 calls pause, never a surprise charge. Authenticate with the Authorization: Bearer header.
| Option | What it does | Where it stops |
|---|---|---|
| Sweepivo AI | Answers calls as a receptionist and plans routes, inside Sweepivo. | Not a connector for ChatGPT, Claude or your own scripts. |
| Sweepivo owner app | The owner home, schedule, payments and payroll on your phone. | You tap through each visit yourself. |
| Imports | Brings data in from spreadsheets, another booking app's export or screenshots. | A way to switch to Sweepivo, not a route for an agent. |
| By hand | Open each screen and copy what you need. | Only while you are at the screen. |
If Sweepivo publishes an API that covers your job, use it: it is the vendor's own route. Cotlet fits when you want ChatGPT, Claude or a script working in Sweepivo today, from your own sign-in, with nothing to host.
Cotlet is run by Take Group sp. z o.o., a software company registered in Poland, EU (VAT PL6342984273) and in business since 2019. Its founders are Bart Krawczyk (Co-founder, CTO) and Jake Slugocki (Co-founder, SEO, sales and marketing). Write to contact@cotlet.dev with questions about the Sweepivo tools.
Sign up free at https://cotlet.dev/signup?app=sweepivo, turn on Sweepivo under Integrations in the console and sign in at https://sweepivo.com/signin in your Cloud Browser. Then connect your AI agent and ask it what needs you today. The trial needs no card.
Sweepivo publishes no public API, developer documentation or API keys (checked 15 Nov 2026). No Sweepivo MCP server and no official ChatGPT or Claude connector was found; Sweepivo AI is the receptionist and route planner inside the product. Cotlet does not need one: it works in your own signed-in Sweepivo account.
Sweepivo signs owners and cleaners in with a 6-digit code sent by email, or a passkey, with an optional authenticator app for two-step sign-in. You enter the code yourself in your Cloud Browser once; the agent then uses that session. Cotlet never sees or stores your password.
The tools see what your plan includes. Payroll comes with Sweepivo's Shine plan and above, so review_cleaner_payroll needs Shine, Sparkle or Spotless. Bookings, scheduling and the cleaner app are in every plan, from Sweep at $29 a month (Sweepivo pricing, checked 15 Nov 2026).
As the owner or a manager. A cleaner login opens the cleaner app: the next job, directions and checklists. Payments, payroll and reports are owner work, so the agent needs an owner or manager session for them.
Sweepivo runs in English and Spanish for owners, cleaners and clients. The tools return what your Sweepivo account shows, and you can ask your agent in either language.
Only in how it is called. Both carry the same 10 tools, inputs, answers, API key and price. Agents use MCP; scripts and n8n HTTP steps use REST.
No. Cotlet is independent and not affiliated with Sweepivo or Ziontech Solutions LLC.
Free 7-day trial with 2,000 credits, no card needed.
May we use Google Analytics to see which pages help people? It sets cookies only if you accept. No ads, and we never sell data. Cookie policy