Code with real bank data in your IDE
Add the BankSync MCP server to Continue and let your open-source coding assistant work with real transactions, balances and investments.
queries/merchant_totals.sql
-- Top merchants, last 30 daysselect merchant, count(*) as charges, sum(-amount) as spentfrom transactionswhere date >= current_date - 30 and amount < 0group by merchantorder by spent desclimit 10;
Agent
Connect Continue 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 the mcpServers list in config.yaml. Headers go under requestOptions.
~/.continue/config.yaml
mcpServers: - name: BankSync type: streamable-http url: https://mcp.banksync.io requestOptions: headers: X-API-Key: bsk_your_key_here - 03
Use agent mode
Switch Continue to agent mode and build with your real data in context.
Write a query for our top merchants and check it against real data.
get_transactions · 214 rows · Business Checking ••4021
merchant_totals.sql is ready. Against the last 30 days it returns AWS ($1,212), Gusto ($8,420) and Figma ($540) at the top.
What you can build.
- Build with real data context
- Generate code that works with actual transaction fields, balance schemas, and account types from your bank.
- Financial app development
- Scaffold budgeting apps, expense trackers, and portfolio dashboards with live bank data in your IDE.
- Data pipeline code
- Write ETL scripts and data pipelines with the AI understanding your bank's actual data structure.
- Test with live data
- Generate test fixtures and mock data based on real transaction patterns from your accounts.
- API integration code
- Let Continue generate API client code with accurate type definitions from BankSync's endpoints.
- Open source flexibility
- Use any LLM provider (OpenAI, Anthropic, local models) through Continue while accessing BankSync data via MCP.
Change assistants. Keep your data.
What Continue 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
- ~/.continue/config.yaml, as a list entry under mcpServers.
- 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.
How do I add BankSync MCP to Continue.dev?
Generate a BankSync API key, then add the MCP server configuration to your .continue/mcpServers/ folder or your Continue JSON config file. The BankSync tools will be available in Continue's agent mode.
Does Continue.dev support remote MCP servers like BankSync?
Yes. Continue.dev now supports remote server connections through HTTP-based transports, enabling integration with cloud-hosted MCP servers like BankSync.
Which IDEs does Continue.dev work with?
Continue.dev is an open-source extension that works with VS Code and JetBrains IDEs (IntelliJ, PyCharm, WebStorm). MCP tools like BankSync are available in agent mode across all supported IDEs.
What can I build with Continue.dev and BankSync?
Build finance tools with live bank data context, generate code that works with real transaction schemas, automate financial data pipelines, and create budget tracking tools, all with AI assistance in your IDE.