Connect Gemini to your bank data
The BankSync MCP server gives Gemini your bank accounts, transactions and balances. It works with Gemini CLI, Google AI Studio and the Gen AI SDK.
Connect Gemini in three steps.
- 01
Create a scoped key
Connect your banks in BankSync, then create an API key with read scopes under Developers.
New API key
NameMCP clientScopesbanks:readfeeds:readKeybsk_••••••••3f9a - 02
Add the server
Add BankSync to Gemini CLI settings, or to the MCP config of your AI Studio or Gen AI SDK app.
~/.gemini/settings.json
{ "mcpServers": { "banksync": { "httpUrl": "https://mcp.banksync.io", "headers": { "X-API-Key": "bsk_your_key_here" } } } } - 03
Ask about your finances
Ask about balances, transactions, spending patterns or holdings.
Summarise our spending by category for October.
get_transactions · 96 rows · Total Checking ••7731
$4,118 in October. Housing $2,100, groceries $642, dining $412 and transport $286 are the biggest.
What you can ask.
- Analyse spending with Gemini
- Ask Gemini to categorise transactions, spot trends, and summarise monthly spending across all your bank accounts.
- Build financial apps with AI Studio
- Prototype AI-powered financial tools inside Google AI Studio using BankSync MCP as a live data source.
- Query bank data from Gemini CLI
- Use the open-source Gemini CLI with built-in MCP support to pull account balances, recent transactions, and institution details from your terminal.
- Combine bank data with BigQuery
- Pair BankSync MCP with Google Cloud's managed MCP servers to blend bank transactions with warehouse analytics in BigQuery.
Change assistants. Keep your data.
What Gemini gets.
- Server
- Remote and hosted at https://mcp.banksync.io over Streamable HTTP. Nothing to install.
- Tools
- 227 documented tools: bank data, feeds and jobs, tables, dashboards, automations and enrichments. Tools reference
- Sign-in
- A scoped API key, sent in the X-API-Key header.
- Config
- Gemini CLI settings, AI Studio’s MCP panel, or your Gen AI SDK app.
- Access
- Set by the scopes you grant. Read scopes keep the connection read-only; no tool can move money.
- Rate limits
- 20 tool calls a minute on Standard, 40 on Professional, 60 on Business, shared across a workspace.
- Banks
- 12,000+ institutions in the US, Canada, the UK, Europe and Australia, including Consumer Data Right connections.
Questions, answered.
Does Google Gemini support MCP servers?
Yes. The Google Gen AI SDK natively supports MCP definitions, the open-source Gemini CLI has built-in MCP support, and Google AI Studio includes MCP demo apps. Google Cloud also provides managed remote MCP servers for services like Maps, BigQuery, and Compute Engine.
How do I connect BankSync to Gemini CLI?
Get a BankSync API key from the dashboard, add the BankSync MCP server configuration to your Gemini CLI settings file with the server URL and your API key in the headers, then start Gemini CLI and ask about your finances.
Can I use BankSync MCP with Google AI Studio?
Yes. Google AI Studio supports MCP servers. Add the BankSync MCP server URL and your API key in the MCP configuration section of AI Studio, and you can prototype financial AI applications with live bank data.
What bank data can Gemini access through BankSync MCP?
Gemini can access account balances, transaction history, institution details, investment holdings, loan information, and more through BankSync's 227 MCP tools. Data comes from 12,000+ supported banks and financial institutions.
Is my bank data secure when using BankSync with Gemini?
Yes. BankSync uses scoped API keys so you control exactly which data each key can access. All communication between Gemini and the BankSync MCP server happens over HTTPS. Bank credentials are never shared with AI models.