Tink MCP for AI. Manage your open banking data and payments in conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
How this MCP server connects to your AI agent
Tink MCP Server lets your agent connect directly to open banking data. Use it to list all connected bank accounts, pull filtered transaction histories by date range, or initiate payment requests—all from simple text commands.
What AI agents can do with Tink Automation
List accounts
Retrieves all bank accounts linked to your user profile.
Create payment request
Sends a payment request to a specific destination account with defined amounts, currencies, and references.
List transactions
Fetches a list of transactions for an account ID over a specified time range.
Gets a list of all bank accounts linked to your profile and retrieves their necessary identifiers.
Retrieves financial transaction records for a specific account ID within a specified date window.
Initiates a payment request, defining the recipient's destination and providing the amount, currency, and reference text.
Ask an AI about this
Waiting for input…
What AI agents can do with Tink: 3 Tools for Financial Operations
These three tools give you full access to open banking capabilities—from listing accounts to sending money.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Tink on VinkiusList Accounts
Retrieves all bank accounts linked to your user profile.
Create Payment Request
Sends a payment request to a specific destination account with defined amounts...
List Transactions
Fetches a list of transactions for an account ID over a specified time range.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Tink integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Tink, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Tink. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 3 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Sifting through bank statements shouldn't mean juggling ten different websites., Solved with Vinkius AI Gateway
Today, checking your finances means logging into a web portal for every single account you own. You click 'Statements,' download a PDF or CSV, open Excel, and then spend 30 minutes just cleaning up the data so it actually makes sense for reporting.
With this MCP server, you simply ask your agent to run `list_transactions`. It pulls the raw data directly into the chat, structured and ready to use. You stop managing web portals and start asking questions.
Tink MCP Server: Initiate payments with a single command.
The old way was logging into your banking app, navigating menus until you hit 'Send Money,' manually entering the IBAN, confirming details on a separate screen, and finally hitting submit. It's tedious friction.
Now, you tell your agent to run `create_payment_request`. You provide the recipient, amount, currency, and reference—all in one prompt. The rest is handled instantly by the server.
What your AI can actually do with this
Forget logging into a dozen banking websites just to check your balance or send some cash over. This server lets your agent connect straight up to open banking data through Tink. You'll manage account details and execute payments using plain text commands—no navigating menus, no clicking buttons. It’s all done right within the chat window you're already in.
To get started, you gotta know what accounts are connected. Your agent uses list_accounts to pull a comprehensive list of every bank account tied to your profile. When it runs this tool, you don't just get names; you get the crucial unique identifiers for each linked account. You need those IDs because they let your agent target specific services later on.
Once you’ve got those identifiers, checking your spending history is easy. The list_transactions tool lets you pull transaction records from a specific bank account ID over any time window you define. You just tell your AI client which account to check and what date range you're interested in—say, 'Show me everything for Account X between January 1st and March 3rd.' The agent runs the tool, and it pulls back every single transaction that falls within those parameters.
This is way better than trying to download a messy CSV file and figuring out which columns mean what.
And when you're ready to move money, create_payment_request handles the whole thing. You don't need to jump through hoops or find routing numbers; you just tell your agent who you're sending the cash to, how much it should be for, which currency you're using, and what reference text needs to go with it.
The tool takes that destination account ID, pairs it with the specific amount, confirms the currency code, and sends the entire request package off. It’s a single command doing three jobs: defining the recipient, setting the value, and triggering the payment process.
Think of the workflow like this: First, you run list_accounts so your agent knows all your options. Then, if you need to review spending, you use those returned account IDs with list_transactions, defining a precise start date and end date for the records you want. If everything looks good—or bad—and you're ready to move funds, you run create_payment_request.
You give it all the pieces: the specific recipient’s destination details, the exact amount in dollars or euros, the currency code, and a clear reference message that explains why the payment is happening. The agent bundles all those inputs and sends out the request immediately.
This whole process means you're not stuck toggling between your AI client and your bank portal. Your agent acts as the middleman, using these three tools to pull data or trigger actions directly based on natural language commands. You talk to your client like you’d talk to a coworker—you tell it what you want done—and it handles the complex API calls behind the scenes.
It's about making financial operations feel less like technical work and more like just talking things through.
019ea60b-8249-71c6-b1fa-cc321a260bf6 
Here's how it actually works
The bottom line is: you talk to your agent, and the agent talks to your bank via Tink.
Subscribe to this server and provide your Tink Access Token.
Your AI client calls the required tool (e.g., list_accounts) using your token.
The server executes the API call, returns structured data, and hands it back to your chat interface.
Who is this actually for?
Anyone who needs reliable financial data access without manual API calls or CSV exports. This targets analysts stuck in reporting loops, developers building payment flows, and business owners needing quick cash flow checks.
Pulls transaction history for multiple accounts across specific date ranges to build reports instantly.
Tests payment initiation logic and account connectivity directly from the code editor environment.
Checks current cash flow status by listing accounts, and initiates necessary transfers using simple text commands.
What Changes When You Connect
Check accounts instantly. Instead of logging into multiple bank portals, just ask your agent to run list_accounts and see every linked account ID immediately.
Build reports faster with list_transactions. Pull transaction records for specific accounts and date ranges without having to manually download and clean CSV files.
Initiate payments via text. Use create_payment_request to send money, defining the recipient, amount, and reference in a single chat turn.
Single source of truth. You manage all your financial data—accounts, transactions, payments—through one reliable connection point.
Cross-platform access. Whether you're using Claude or Cursor, your agent handles the complex bank API calls, keeping your workflow simple.
See it in action
The Q3 Expense Audit
A financial analyst needs to audit all spending from a specific account last quarter. They ask their agent: 'Show me transactions for acc_1234 from July 1 to Sept 30.' The agent runs list_transactions, providing the full, filtered list instantly for report building.
The Payroll Check
An operations manager needs to confirm available funds before running payroll. They ask their agent to run list_accounts. Seeing the balance confirms they have enough cash and can then use create_payment_request immediately.
The Developer Test Flow
A developer is building a payment microservice. Instead of setting up mock endpoints, they ask their agent to run list_accounts and then simulate a transfer using create_payment_request, testing the full flow in real-time.
The New Account Check
A business owner just opened a new corporate account. They ask their agent to run list_accounts. The server immediately lists the new ID, allowing them to use it for future payments or reporting right away.
The honest tradeoffs
Manually downloading reports
Logging into Bank A's portal, exporting a CSV. Logging into Bank B's portal, exporting another CSV. Manually merging and filtering the data in Excel.
Tell your agent to run list_transactions for both accounts and the required date range. The server handles the aggregation and gives you clean, structured data.
Forgetting account IDs
Trying to view transactions without first knowing which specific internal Account ID (e.g., acc_8821) belongs to 'Main Checking'. The API call fails or returns garbage.
Always run list_accounts first. This provides the necessary IDs you must reference when calling list_transactions.
Sending payments without checking balances
Attempting to use create_payment_request for a large sum before confirming the account has sufficient funds, leading to failed transactions or errors.
Check the balance and transaction history using list_accounts and list_transactions before you call create_payment_request. This confirms state integrity.
When It Fits, When It Doesn't
Use this server if your core workflow involves checking account balances, viewing historical transactions, or initiating payments. The moment your process requires reading financial data or moving money, this is the tool stack you need.
Don't use it if: You only need to read unstructured text from a document, manage a simple database record (like user preferences), or send non-financial messages. For those tasks, look for dedicated messaging or CRM tools. If your goal is pure data processing without financial context, a general data connector will do better. But when money and accounts are involved? This is the one.
Questions you might have
How do I use Tink MCP Server to see all my bank accounts? +
You run the list_accounts tool. This returns a list of every account connected to your profile, giving you the necessary IDs for other tools.
Can I get transactions from multiple banks using Tink MCP Server? +
Yes, you first use list_accounts to identify all relevant accounts. Then, you run list_transactions separately for each account ID and combine the results in your analysis.
What data does create_payment_request require? +
It requires the destination account details (IBAN/ID), the exact amount, the currency code, and a clear reference for tracking purposes.
Does Tink MCP Server handle transaction filtering by date? +
Yes. When using list_transactions, you specify both an account ID and a start/end date range to narrow down the records.
How do I handle an expired token when using the Tink MCP Server for `list_accounts`? +
The server requires a valid, current Tink Access Token. If your token expires, you must generate a new one through Tink and update it in Vinkius. This keeps the connection active and prevents account listing failures.
What does `create_payment_request` do if the recipient IBAN is invalid? +
It returns an explicit error code detailing the validation failure. Your AI client receives this specific feedback, letting you know immediately that the payment request needs a corrected destination account number.
Can `list_transactions` filter by transaction type (e.g., debit or credit)? +
Yes, in addition to date ranges and accounts, you can specify filters for activity types. This lets your agent pull only 'debit' entries, for instance, making deep analysis faster.
Are there any rate limits when calling `list_transactions` repeatedly? +
While Vinkius manages the connection, Tink enforces API limits on high-volume requests. Running too many queries in quick succession will result in a 429 error, and you'll need to slow your AI client down.
How can I see all my connected bank accounts at once? +
You can use the list_accounts tool. It will retrieve a complete list of all bank accounts currently linked to your Tink profile.
Is it possible to filter my spending by specific dates? +
Yes! The list_transactions tool allows you to provide an optional startDate and endDate (ISO 8601) to narrow down your transaction history.
Can I initiate a bank transfer using this server? +
Yes, you can use the create_payment_request tool. You will need to provide the destination account number, type, amount, and recipient details to start the process.
We've already built the connector for Tink. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 3 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.