Skip to main content
Learn how Grand Central exposes banking operations as discoverable MCP tools.

​
Tool discovery

Your AI agent can discover all available tools through the MCP protocol: 
{
  "jsonrpc": "2.0",
  "method": "tools/list",
  "id": 1
}
Response includes tool metadata: 
{
  "jsonrpc": "2.0",
  "result": {
    "tools": [
      {
        "name": "getCustomerProfile",
        "description": "Retrieve customer contact information and preferences",
        "inputSchema": {
          "type": "object",
          "properties": {
            "customerId": {
              "type": "string",
              "description": "Unique customer identifier"
            }
          },
          "required": ["customerId"]
        }
      }
    ]
  },
  "id": 1
}

​
Tool metadata

Each tool provides complete metadata for AI agents:

Name

Unique identifier for the tool (e.g., getCustomerProfile)

Description

Clear explanation of what the tool does and when to use it

Input schema

JSON Schema defining required and optional parameters

Examples

Sample requests showing proper usage

​
Tool organization

Grand Central organizes tools by business domain. Your available tools depend on which domains are enabled for your subscription:

Payments

Payment processing, transfers, fees, status tracking

Accounts

Account balances, transactions, details, statements

Customers

Customer profiles, contact information, preferences

Loans

Loan eligibility, applications, interest rates

Cards

Card management, limits, transactions

Products

Available banking products, terms, conditions
Available domains and tools depend on what you’ve enabled in the admin portal for your subscription.

​
Adding new tools

Need access to additional banking operations? Log into the admin portal and enable them yourself. For operations already in the catalog: Browse the API catalog in the admin portal’s MCP Tools section, select the operations you need, configure rate limits and authentication requirements, and enable them. Pre-configured operations are available immediately. The system runs automated validation to ensure security constraints are met. For custom operations not yet in the catalog: Upload or link your OpenAPI specification through the portal. Select which operations to expose as MCP tools, and the system runs automated security validation. Low-risk operations (read-only, public data) enable immediately and appear in discovery responses within minutes. Higher-risk operations (write access, PII exposure) get flagged for automated compliance scanning - these typically complete within 1-3 business days. Platform support is available if you need help with complex specifications or have questions about security requirements.

​
Best practices

Cache tool list

Cache discovery responses for 5-10 minutes to reduce API calls

Validate parameters

Use the input schema to validate before invoking tools

Handle errors gracefully

Implement retry logic with exponential backoff

Monitor usage

Track which tools are used most frequently

​
Next steps

Tool invocation

Learn how to invoke tools

Authentication

Understand authentication requirements