Build Harvest-Powered Applications in Gemini Code Assist with CData MCP Server

Gemini Code Assist is an AI-powered coding companion that integrates intelligent code generation into everyday development workflows. With support for MCP, Gemini Code Assist can connect to live enterprise data sources directly from Visual Studio Code, enabling natural language interaction with structured data without switching context or manually writing data access code.

Model Context Protocol (MCP) is an open standard for connecting LLM clients to external services through structured tool interfaces. MCP servers expose capabilities such as schema discovery and live querying, allowing AI agents to retrieve and reason over real-time data safely and consistently.

This guide explains how to install the CData API Driver for MCP Server, configure the connection to Harvest, connect the MCP Server add-on to Gemini Code Assist, and query live Harvest data from within the editor.

Prerequisites

Before starting, ensure the following requirements are met:

1. Visual Studio Code is installed on the machine
2. Gemini Code Assist extension is enabled in Visual Studio Code
3. CData API Driver for MCP Server has been installed
4. Access to Harvest

Note: Gemini Code Assist must already be set up and functional in Visual Studio Code before configuring MCP servers. MCP servers are accessed when Gemini Code Assist is running in Agent mode.

Step 1: Download and install the CData API Driver for MCP Server

1. To begin, download the CData API Driver for MCP Server
2. Find and double-click the installer to begin the installation
3. Run the installer and follow the prompts to complete the installation

When the installation is complete, the MCP Server add-on is ready for configuration by connecting to Harvest.

Step 2: Configure the connection to Harvest

1. After installation, open the CData API Driver for MCP Server configuration wizard

   NOTE: If the wizard does not open automatically, search for "CData API Driver for MCP Server" in the Windows search bar and open the application.

   
2. In MCP Configuration > Configuration Name, either select an existing configuration or choose to create a new one
3. Name the configuration (e.g. "cdata_api") and click OK

4. Enter the appropriate connection properties in the configuration wizard

   Start by setting the Profile connection property to the location of the Harvest Profile on disk (e.g. C:\profiles\Harvest.apip). Next, set the ProfileSettings connection property to the connection string for Harvest (see below).

   Harvest API Profile Settings

   To authenticate to Harvest, you can use either Token authentication or the OAuth standard. Use Basic authentication to connect to your own data. Use OAuth to allow other users to connect to their data.

   Using Token Authentication

   To use Token Authentication, set the APIKey to your Harvest Personal Access Token in the ProfileSettings connection property. In addition to APIKey, set your AccountId in ProfileSettings to connect.

   Using OAuth Authentication

   First, register an OAuth2 application with Harvest. The application can be created from the "Developers" section of Harvest ID.

   After setting the following connection properties, you are ready to connect:

   • ProfileSettings: Set your AccountId in ProfileSettings.
   • AuthScheme: Set this to OAuth.
   • OAuthClientId: Set this to the client ID that you specified in your app settings.
   • OAuthClientSecret: Set this to the client secret that you specified in your app settings.
   • CallbackURL: Set this to the Redirect URI that you specified in your app settings.
   • InitiateOAuth: Set this to GETANDREFRESH. You can use InitiateOAuth to manage how the driver obtains and refreshes the OAuthAccessToken.
   
5. Click Connect to authenticate with Harvest through OAuth
6. Then, click Save Configuration to save the MCP Server add-on

This process creates a .mcp configuration file that Gemini Code Assist will reference when launching the MCP Server add-on. With the MCP Server add-on configured, it is ready to connect to Gemini Code Assist.

Step 3: Connect the MCP Server add-on to Gemini Code Assist

1. Ensure Visual Studio Code is installed and the Gemini Code Assist extension is enabled
2. From the configuration wizard, click Next after saving and testing the connection
3. Select Gemini Code Assist from the AI MCP Tool dropdown
4. Click Copy JSON to copy the generated MCP configuration to the clipboard
5. Paste the copied JSON into the appropriate configuration file based on the desired scope:
   • User-level: Configuration applies across all projects for the current user
   • Workspace-level: Configuration applies only to the current workspace or project

   NOTE: The configuration includes the path to Java 17+ executable and the CData MCP Server add-on JAR file. The final argument must match the MCP configuration name saved in the wizard (e.g. "cdata_api").

6. Save the configuration file and restart Visual Studio Code if necessary

Step 4: Query live Harvest data in Gemini Code Assist

1. Open Visual Studio Code and select Gemini Code Assist in the activity bar
2. Enter /mcp in the chat prompt to verify the connection status. The Harvest MCP Server add-on should appear with a green connection indicator
3. Ask questions about Harvest data using natural language. For example:

   "Provide the list of all tables available in my Harvest data connection."

   
4. Generate code that works with live Harvest data. For example:

   "Write a function to retrieve records from the Invoices table where Id matches a given value."

   

Gemini Code Assist is now fully integrated with the CData MCP Server add-on and can use the MCP tools exposed to explore schemas, execute live queries against Harvest, and generate data-aware code.

Build with MCP Server. Deploy with CData Drivers.

Download MCP Server for free and give your AI tools schema-aware access to live Harvest data during development. When you're ready to move to production, CData Harvest Drivers deliver the same SQL-based access with enterprise-grade performance, security, and reliability.

Visit the CData Community to share insights, ask questions, and explore what's possible with MCP-powered AI workflows.