YNAB MCP: Community Servers, No Official One
No — YNAB (You Need A Budget) does not have an official MCP server. YNAB hasn’t shipped anything for the Model Context Protocol itself. What it does have is a well-documented, free public API — and a handful of community developers have wrapped that API in MCP servers you can run yourself. YNAB’s own developer docs even list one of them, a “Local MCP Server for YNAB,” under community-built tools, but it’s the community’s work, not YNAB’s own connector.
Even so, the same limit applies as with any MCP server: it hands YNAB to an AI inside a conversation you start. It’s a doorway, not a worker. Nothing watches your budget for you, nothing fires when a category goes over or a big transaction lands, and nothing runs while the chat is closed. Here’s the honest state of YNAB and MCP, what the community servers actually do, where the limits are — and what to use when you want YNAB work that runs on its own.
For the wider picture, every app with an MCP server is listed by category.
What an MCP connection to YNAB does
Model Context Protocol (MCP) is the open standard that lets an AI client — Claude, ChatGPT, Cursor, and others — talk to an outside app through a shared interface. YNAB hasn’t built one. The servers people mean when they say “YNAB MCP” are independent projects — calebl/ynab-mcp-server, klauern/mcp-ynab, jeangnc/ynab-mcp-server, and EthanKang1/ynab-mcp, among several others — each calling YNAB’s own REST API under the hood using a personal access token you generate yourself.
With one connected, an AI client can typically:
- Read your budget and accounts — list your budgets, pull account balances, and see category balances and how much you’ve assigned versus spent.
- Pull transactions — fetch recent or unapproved transactions from an account so the AI can summarize spending or flag something unusual.
- Create and categorize transactions (on most servers) — add a transaction, assign it to a category, or approve a pending one, using your own account’s access token.
It’s genuinely useful for ad-hoc questions — “which categories am I overspending this month, and by how much?” — answered from your live budget instead of a training-data guess. Some servers, like jeangnc/ynab-mcp-server, even track their write operations so you can undo them.
How to connect YNAB to AI in 2026
There’s no “install the official server” step, because YNAB didn’t build one. The realistic path is:
- Generate a YNAB personal access token. In YNAB, open Account Settings → Developer Settings and create a new token at api.ynab.com. This is what every community YNAB MCP server authenticates with — you’re acting as your own account owner, not through any YNAB blessing of the MCP project.
- Know the API’s limits. The YNAB API is free to use with your existing YNAB subscription, but it’s capped at 200 requests per hour per access token in a rolling window. That’s plenty for asking questions, but it matters if a workflow polls your budget often.
- Add the community MCP server to your AI client, set your personal access token as its environment variable, and restart the client so the tools register.
- Test with a read query first — “list my budgets” or “show this month’s category balances” — before trusting any server with transaction-creating tools, since those write to your real budget.
For most people searching “connect YNAB to AI,” this is a real setup with a real token and rate limit attached, not a one-click connector.
Where the YNAB MCP stops
Even setting the “which community server do I trust” question aside, the same four limits show up here:
- It only works inside a chat you start. Close the window and nothing happens. Nothing is watching your budget or your accounts for you.
- No triggers. A category going negative, a large transaction importing, the first of the month arriving with money to assign — none of these can start anything through MCP. There’s no “when this happens in YNAB, do that.”
- It’s one app at a time. A YNAB MCP server knows YNAB. Getting an overspending alert into Slack, a spending summary into a Google Sheet, or a bill reminder onto your calendar means wiring up a separate MCP server for each and hoping your client can juggle them in one turn.
- You own the plumbing, the token, and the account risk. There’s no vendor support line here — you generated the token, you’re metered against the 200-per-hour limit, and a server with transaction tools writes to your real budget with no one else reviewing what it does.
So a YNAB MCP server, community-built as all of them are, is not a way to make YNAB run — to have work happen on a schedule or in reaction to an event, across the other tools your money touches.
Running YNAB work that doesn’t need a chat open
That “run on its own, across apps” gap is exactly where Carly fits. Carly connects to YNAB and to the ~260 other apps it supports natively, plus anything with a public API through your own key — no community MCP server to vet, no separate token to wire into a client yourself. The difference from MCP is the important part: Carly’s workflows are triggered and scheduled, so YNAB work happens whether or not anyone has a chat window open.
A few things a Carly workflow can do that no MCP server can:
- When a category’s balance goes negative → post the overspending to a Slack channel and note which transaction pushed it over.
- Every morning → summarize yesterday’s transactions and this month’s category balances, and email you the digest.
- On the 1st of each month → pull the prior month’s spending by category into a Google Sheet and flag any category that ran over its assigned amount.
The non-AI steps — the moving, matching, and routing between apps — are free and unlimited, the Zapier-style backbone of the workflow. The AI steps (drafting, summarizing, deciding) start at $35/month. You describe the outcome in plain language and Carly wires up the YNAB connection and everything downstream.
If you just want to ask questions about your budget from a chat, a community MCP server is the closest thing available, with a YNAB personal access token and its rate limit behind it. If you want YNAB activity to actually trigger something — on a schedule, in reaction to an event, across every app your money flows through — that’s the job no YNAB MCP server was built for.
FAQ
Does YNAB have an official MCP server?
No. YNAB hasn’t published one. What exists are community projects — such as calebl/ynab-mcp-server, klauern/mcp-ynab, and jeangnc/ynab-mcp-server — that call YNAB’s own API through a personal access token you generate yourself. YNAB’s developer docs list one of these community servers, but it isn’t YNAB’s own connector.
Is the YNAB API free to use for MCP? Yes, the API is free to use with your existing YNAB subscription — there’s no per-call billing. It is rate-limited to 200 requests per hour per access token, so a YNAB MCP server that polls your budget frequently can hit that ceiling.
Can a YNAB MCP server trigger automations? No. MCP is request/response inside an AI chat — it has no triggers and nothing runs when the conversation is closed. For event- or schedule-driven YNAB work across apps, you need a workflow tool like Carly rather than an MCP server.
Can I connect YNAB to AI without coding or hosting a server? Yes. You don’t have to generate a token or run an MCP server at all. Carly connects to YNAB for you and lets you build the automation in plain language — describe what you want to happen and it wires up the YNAB connection and the other apps involved, with no server to host and no code to write.
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."
