ChatGPT + Expensify: What the Integration Can (and Can't) Do in 2026
Yes, ChatGPT can connect to Expensify — not through a listed app in ChatGPT’s directory, but through Expensify’s official MCP server, which you add as a custom connector in about five minutes. Once connected over OAuth, you can interrogate your live expense data in plain English: what you spent on flights last quarter, which reports are stuck unapproved, whether that Marriott receipt ever got attached. Two things to know up front. First, Expensify built the connector read-only by design — it can analyze everything, but it cannot approve reports, edit expenses, or move money. Second, like every connector in ChatGPT, it works in a session you’re driving — between chats, nothing watches your reports or receipts.
Here’s what the ChatGPT Expensify integration actually does, how to set it up, and what to use when you want expense work that runs without you.
What ChatGPT can actually do with Expensify
- Query expenses, receipts, reports, and invoices live. “How much did I spend on Uber and Delta in Q2?” gets answered from your real account data, not an export.
- Chase down report status. “Which of my reports are still unsubmitted or waiting on approval?” — without clicking through the Expensify UI.
- Audit receipts conversationally. Ask which expenses over $75 are missing receipts, then narrow by date, merchant, or category in follow-ups.
- Analyze spend patterns. Month-over-month travel spend, top merchants, category breakdowns — the kind of thing that used to mean a CSV export and a pivot table.
- Run inside agent sessions. With ChatGPT Work (launched July 9, 2026), you can @-mention connected apps and let an agent work across Expensify and the rest of your connected stack for a long, metered run — a quarterly T&E deep-dive, say. Still a run you start.
How to set it up
- In ChatGPT, open Settings → Apps → Advanced settings and enable Developer mode (required for custom MCP connectors).
- Back in Settings → Apps, choose Create app. Name it Expensify and set the connection URL to
https://www.expensify.com/mcp. - Click Create, then complete the OAuth sign-in in your browser — no API keys, exports, or admin setup needed on the Expensify side.
- Ask something (“show me my unsubmitted expenses”) or invoke it explicitly with @Expensify in a prompt.
The limits that actually matter
- Read-only, full stop. Expensify is explicit that the connector can’t approve transactions, edit data, or move money. That’s the right call for a finance tool — but it means every insight ends with you switching to Expensify to act on it.
- It doesn’t run on triggers. There’s no “when a report is submitted, check it against policy” or “when an expense over $500 posts, flag it.” ChatGPT queries Expensify when you prompt it — it never fires on an Expensify event.
- Session-bound, even in agent mode. ChatGPT Work runs are long and autonomous but manually started and metered against your plan’s allowance — an errand, not a standing watch on your spend.
- Cross-stack follow-through stops at the chat. It won’t email the employee whose receipt is missing, ping an approver in Slack, or log a summary to a spreadsheet on a schedule.
If you want Expensify work that runs on its own: Carly
The moment you want something to happen off an expense event — a policy check the instant a report is submitted, a Friday-afternoon reminder to everyone sitting on unsubmitted expenses, a flag when card spend at a single merchant spikes — you’ve crossed past what a read-only connector in a chat is for.
That’s where Carly fits. Carly is an AI executive assistant that acts on triggers across your whole stack, set up by conversation instead of code:
- Fires on events and schedules, 24/7, in the cloud. Report submitted, threshold crossed, Friday 3pm — Carly acts without a chat open.
- No-code setup. Tell Carly “when an expense report is submitted, check it against our policy and ping the approver” in plain English; it interviews you and builds the workflow.
- Connects expenses to the rest of your work — Expensify data flowing into email, Slack, spreadsheets, and your accounting stack in one flow.
- Actually sends — drafts and sends email across Gmail and Outlook, posts to Slack, updates records, manages tasks.
- Connects to anything — 200+ native integrations, plus any other tool via your own API key. Carly connects to Expensify with your own API key — paste it on carlyassistant.com/integrations.
AI agents start at $35/month, and steps in a workflow that don’t use AI run free and unlimited. See integrations for the full list.
ChatGPT vs Carly
| ChatGPT (Expensify MCP) | Carly | |
|---|---|---|
| Query live expenses, reports, receipts | Yes | Yes |
| Conversational spend analysis | Yes | Yes |
| Friday unsubmitted-expense reminders, unprompted | No | Yes, on a schedule |
| Reacts to a submitted report by itself | No | Yes, on any trigger |
| Runs without a session open | No (agent runs are started + metered) | Yes (cloud, 24/7) |
| Pushes findings into email / Slack / sheets | No | Yes |
| Emails the reminder to the employee | No | Yes (Gmail + Outlook) |
| Setup | Developer mode + custom connector | Describe it in plain English |
| Pricing | Paid ChatGPT plan | AI agents from $35/mo |
ChatGPT’s Expensify connector is an analyst you question in a chat. Carly is an assistant that acts on your expense data while you’re doing something else.
Frequently Asked Questions
Does ChatGPT work with Expensify?
Yes. Expensify ships an official hosted MCP server at expensify.com/mcp. Turn on Developer mode in ChatGPT’s app settings, add it as a custom app, and authorize via OAuth. You can then query expenses, receipts, reports, and invoices from your live account in plain English.
Can ChatGPT approve expense reports or edit expenses in Expensify?
No. Expensify built the connector read-only by design — AI assistants can read and analyze the same real-time data you see in Expensify, but they cannot approve transactions, edit data, or move money. Any action still happens inside Expensify itself.
Can ChatGPT react to a new expense report automatically?
No. ChatGPT queries Expensify inside a session you start — it doesn’t watch for submitted reports, missing receipts, or policy violations. For “when a report is submitted, check it and ping the approver,” you need a trigger-based assistant like Carly.
Is there an official Expensify app in ChatGPT’s directory?
Not a listed one as of July 2026 — the official path is Expensify’s own MCP server added as a custom connector in ChatGPT’s Developer mode. It’s Expensify-built and OAuth-secured, just not a one-click directory install.
More: ChatGPT email assistant · ChatGPT personal assistant · ChatGPT MCP · Can ChatGPT send emails · Claude + Expensify · Best AI tools for consultants
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."
