Jobber AI: How to Connect Jobber to an AI Assistant
Yes. You can connect Jobber to AI today, for your own account, without asking anyone’s permission. Jobber’s developer documentation leads with language about approved partners and its App Marketplace, which reads like a locked door. It isn’t one for a contractor connecting their own business.
Once Carly is connected, the quotes you sent last week stop going cold, completed jobs turn into invoices without you sitting down at night, and clients hear from you before they chase you.
What changes in your Jobber account
Your quotes stop going cold. You quote on Monday. By Thursday it’s gone quiet, and in residential trades most of the window where you’d still have won that job has already closed. Carly follows up at day three with the quote attached, tells you if the client raises a concern, and stops the moment the quote is accepted or declined.
Finished jobs invoice themselves. The job completes, the invoice goes out, the client gets a wrap-up with the photos, and a review request follows a few days later. That whole sequence normally happens at 9pm, or not at all.
You get paid sooner. Invoices age past 30 days, reminders go out and escalate on your schedule, and everything stops the second payment lands.
New clients get the same start every time. A client is created in Jobber, the welcome goes out, and the record reaches your CRM and accounting system without double entry.
Your day starts with a brief. Quotes going cold, jobs completing today, unpaid invoices, and tomorrow’s confirmations. One message before you’re in the truck.
Clients stop calling to ask. They already got the update, written from what’s actually in Jobber.
You don’t need approval, and here’s the proof
Jobber says it outright:
If you don’t intend to get your app published in Jobber’s App Marketplace and there is no need to connect it to more than 5 paying Jobber accounts then the app can remain in Draft state. Requesting an app review will not be required.
Connecting your own Jobber account is one paying account. You’re four under the ceiling and no review is required. The partner language in the docs is aimed at companies publishing apps to other Jobber users, not at you automating your own business.
Why Jobber’s data needs something that watches it
Jobber has shipped AI features around quoting and client communication, and they help. There’s no first-party Jobber MCP server, and the small community ones share the same ceiling anyway.
MCP servers wait to be asked. The protocol has no scheduler and no event mechanism — its maintainers acknowledge that change detection today means polling — so an MCP connection can answer a question but cannot notice something on its own.
So a chat assistant tells you which quotes are open when you ask it. That was never the hard part. The hard part is that a quote goes quiet on day five and nobody is watching on day five.
Jobber already knows when a quote is created, when a job is completed, and when an invoice goes out. What’s been missing is something that acts on it.
Everything Carly can reach in Jobber
Jobber’s API exposes your whole operation, not a slice of it. The objects, straight from Jobber’s own list:
Client · Quote · Job · Invoice · Request · Property · Assessment · Visit · Expense · Product or Service · Time Sheet Entry · User · Account
So Carly can read a quote and its line items, look up the property and the client history behind it, check which visits are scheduled, see what’s been invoiced and what’s been paid, and pull the timesheet entries against a job. Then act on all of it.
And it reacts in real time. Jobber’s webhook topics follow an OBJECT_ACTION pattern across those objects, so client created, quote created, job completed, and invoice updated are all live signals. Deliveries are signed with HMAC-SHA256, so Carly can verify every event genuinely came from Jobber.
One detail that explains why generic tools struggle here: Jobber’s API is GraphQL, not REST. There’s a single endpoint and no resource paths, so anything expecting a normal REST URL has nothing to point at. Carly speaks it properly.
Connecting takes about five minutes
- Create an app in Jobber’s Developer Center and leave it in Draft.
- Go to carlyassistant.com/integrations, open Custom OAuth, and paste the link to Jobber’s API docs. Carly fills in the authorization URL, token URL, and scopes for you.
- Paste the client ID and secret, connect, and authorize on Jobber’s screen.
Jobber access tokens expire every 60 minutes, and mishandled refresh tokens are one of the most common causes of integrations dying in production. Carly handles the refresh and rotation, so the connection keeps working without you thinking about it.
Connections are read-only until you approve write access, path by path. Credentials are encrypted and never exposed to AI-generated code.
Ask Jobber to run these
When a quote has been sent but not accepted after three days, send a useful follow-up with the quote attached, notify me if the client replies with a concern, and stop the sequence the moment the quote status changes.
When a job is marked complete, send the client a wrap-up with the photos from that job, create the invoice, and schedule a review request for four days later.
Every morning, tell me which quotes are going cold, which jobs finish today, and which invoices are past terms.
Carly runs these on demand, on a schedule, or the moment Jobber fires the event. AI agents start at $35/month, and workflow steps that don’t use AI run free, which covers most of a Jobber workflow since pulling a client record and sending a message need no model.
And these actually send. Assistant-style AI usually drafts and stops, which means the follow-up sits in a drafts folder. Carly sends through Gmail and Outlook, attachments included.
Rather not create the Developer Center app yourself? Book a call and we’ll set it up with you.
Frequently Asked Questions
Does Jobber have AI?
Jobber has shipped AI features aimed at quoting and client communication. For AI that acts across your whole business, you connect Jobber’s API to an assistant like Carly, which can then chase quotes, invoice completed jobs, and send client updates on its own.
Can I connect Jobber to an AI assistant?
Yes, and for your own account you don’t need approval. An app in Draft state connects to up to 5 paying Jobber accounts with no review required, and connecting your own business uses one of those five.
Do I need Jobber’s permission to use the API?
Not for your own account. Jobber’s docs lead with partner and App Marketplace language, but that applies to publishing an app to other Jobber users. Past 5 paying accounts you’d need approval, which never comes up when you’re automating your own business.
What can AI do with my Jobber data?
The valuable uses are time-sensitive: following up on quotes before they go cold, turning completed jobs into invoices and client wrap-ups, chasing invoices past terms, onboarding new clients consistently, and briefing you each morning on what needs attention.
Does Jobber have an MCP server?
Not a first-party one. A few very small community MCP servers exist, plus middleware options. All of them answer questions inside a conversation you start, so nothing is watching your account while you’re on a job.
Will AI actually send messages to my clients, or just draft them?
That depends on the tool, so check before you buy. Many assistants draft and stop. Carly sends through Gmail and Outlook with attachments, so a quote follow-up or job wrap-up actually goes out.
More: Construction AI · Jobber vs Housecall Pro · Claude + Jobber
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
