Integrating Claude Code with JSON Services via CData Connect AI

Claude Code is an AI-powered development environment that brings intelligent code generation, automation, and interactive reasoning directly into your workflow. By integrating it with CData Connect AI, you can enable Claude Code to securely access, query, and interact with live enterprise data, such as JSON, through a standardized MCP tool interface.

CData Connect AI is a managed MCP platform that exposes your enterprise data sources through the Model Context Protocol (MCP). This allows Claude Code to work with catalogs, schemas, tables, metadata, and SQL-enabled data access from hundreds of data sources, without requiring ETL pipelines or custom integration code.

This article explains how to register the CData Connect AI MCP endpoint in Claude Code, configure your JSON or other data source connection, and begin issuing real-time data queries directly from the coding environment. We explore how Claude Code uses the built-in MCP tools, such as getCatalogs, getSchemas, getTables, and queryData to help you write, debug, and automate development workflows powered by live JSON services securely and interactively.

Prerequisites

• An account in CData Connect AI
• A Claude Code account.
• Visual Studio Code installed on your system.

Step 1: Configure JSON connectivity for Claude Code

For Claude Code to access JSON, create a connection to JSON in CData Connect AI. This connection is then exposed to Claude Code using the remote MCP server.

1. Log in to Connect AI click Sources, and then click + Add Connection
2. From the available data sources, choose JSON
3. Enter the necessary authentication properties to connect to JSON

   See the Getting Started chapter in the data provider documentation to authenticate to your data source: The data provider models JSON APIs as bidirectional database tables and JSON files as read-only views (local files, files stored on popular cloud services, and FTP servers). The major authentication schemes are supported, including HTTP Basic, Digest, NTLM, OAuth, and FTP. See the Getting Started chapter in the data provider documentation for authentication guides.

   After setting the URI and providing any authentication values, set DataModel to more closely match the data representation to the structure of your data.

   The DataModel property is the controlling property over how your data is represented into tables and toggles the following basic configurations.

   • Document (default): Model a top-level, document view of your JSON data. The data provider returns nested elements as aggregates of data.
   • FlattenedDocuments: Implicitly join nested documents and their parents into a single table.
   • Relational: Return individual, related tables from hierarchical data. The tables contain a primary key and a foreign key that links to the parent document.

   See the Modeling JSON Data chapter for more information on configuring the relational representation. You will also find the sample data used in the following examples. The data includes entries for people, the cars they own, and various maintenance services performed on those cars.

   
4. Click Save & Test
5. Once authenticated, open the Permissions tab in the JSON connection and configure user-based permissions as required

Generate a Personal Access Token (PAT)

Claude Code authenticates to Connect AI using an account email and a Personal Access Token (PAT). Creating separate PATs for each integration is recommended to maintain access control granularity.

1. In Connect AI, select the Gear icon in the top-right to open Settings
2. Under Access Tokens, select Create PAT
3. Provide a descriptive name for the token and select Create
4. Copy the token and store it securely. The PAT will only be visible during creation

With the JSON connection configured and a PAT generated, Claude Code is prepared to connect to JSON services through the CData MCP server.

Step 2: Install Claude Code

Claude Code is distributed as an npm package. You can install it globally.

To install Claude Code on your system, open PowerShell, Terminal, or CMD as an Administrator and run:

npm install -g @anthropic-ai/claude-code

Verify the installation using the following command:

npm list -g @anthropic-ai/claude-code

Expected output should be:

C:\Users\User\AppData\Roaming
pm
`-- @anthropic-ai/[email protected]

Step 3: Authenticate Claude Code with Claude.ai

Link your local Claude Code environment with your Claude.ai account to enable secure access. In the terminal, run:

claude login

Claude Code outputs a URL, like:

Please visit https://claude.ai/login?code=

Follow these steps:

1. Click the URL or paste it into your browser.
2. Log in to Claude.ai.
3. Claude.ai displays a verification code.
4. Return to your terminal and enter/paste the provided verification code when prompted.

Once verified, you'll need to authenticate with Claude Code using an authentication code. Once done, your terminal should display:

You're all set up for Claude Code.

Claude Code is now linked to your Claude.ai account.

Step 4: Create a Claude Code project

To set up a workspace where Claude Code can store MCP configuration files, start by creating a new directory:

mkdir ClaudeCode
cd ClaudeCode

Now, open it in Visual Studio Code:

code .

Step 5: Launch Claude Code and register the CData Connect AI MCP server

Before Claude Code can interact with JSON, you must register your CData Connect AI MCP endpoint. Claude Code uses this remote MCP server to securely access metadata, schemas, tables, and live query results.

Now register the CData Connect AI MCP server by running the following command in your Claude Code project directory:

claude mcp add connectmcp https://mcp.cloud.cdata.com/mcp \
  --transport http \
  --header "Authorization: Basic base64encoded(EMAIL:PAT)" \
  --header "Content-Type: application/json"

Once added, verify that Claude recognizes your MCP server:

claude mcp list

If successful, you should see:

connectmcp: https://mcp.cloud.cdata.com/mcp (HTTP) - ✓ OK

Start the Claude Code assistant and verify that it detects your MCP server. To run, use the given command:

claude

Once Claude Code loads, you should see:

Loaded MCP Server: connectmcp

This confirms that Claude Code is now connected to your CData Connect AI instance.

Step 6: Explore JSON metadata

You can now use Claude Code's natural-language interface to list catalogs, schemas, and tables in JSON. Ask Claude:

List all JSON catalogs using getCatalogs.

Claude automatically calls the appropriate MCP tool when you issue a request.

Try additional queries such as:

• "Show the available schemas."
• "List all tables in the JSON connection."
• "Retrieve the top 10 records from the Account table."

Claude Code uses the following MCP tools to interact with JSON in real time:

• getCatalogs
• getSchemas
• getTables
• queryData

These tools allow Claude Code to retrieve metadata and query live JSON services.

Step 7: Generate code and automation workflows

Use real JSON metadata to build working scripts directly inside your IDE.

Example prompt:

Write a Python script that queries Salesforce Contacts where LastName starts with 'A' using the MCP queryData tool.

Claude Code writes accurate code because it has:

• direct access to JSON schemas
• live query testing
• metadata introspection

All delivered through CData Connect AI.

Step 8: Build data-driven development workflows

Use Claude Code to generate, refine, and automate code that works with your JSON services using CData Connect AI.

With the CData Connect AI integration in place, Claude Code can help you build development workflows that rely on your JSON services. Although Claude Code does not include built-in real-time data connectivity, your configured MCP connection through CData Connect AI provides it with access to the metadata and query results for your request.

You can use Claude Code to automate tasks such as:

• generating scripts for data exploration
• creating integration test scaffolding
• validating queries against your JSON schema
• producing code for data extraction or transformation workflows

In this setup, Claude Code acts as an intelligent coding assistant that uses live JSON services from CData Connect AI to help you write and refine data-driven logic.

Optional: Manage MCP integrations

Add, remove, or inspect MCP servers in your project.

List MCP servers using the following command:

claude mcp list

To remove one, use:

claude mcp remove connectmcp

Modify the config by editing:

.claude/mcp.json

Get CData Connect AI

To get live data access to hundreds of SaaS, Big Data, and NoSQL sources directly from your cloud applications, try CData Connect AI today!