Verodat MCP Server

🚀 Verodat MCP Server Version

Verodat MCP Server is a tool for interacting with the Verodat platform. It supports various operations, including account and workspace management, dataset operations, and AI operations. The server is written in TypeScript and comes with detailed documentation and community support.

✨ Features

Tool Classification

1. Consume (Read Operations)

• get-accounts: List accessible accounts.
• get-workspaces: List workspaces within an account.
• get-queries: Retrieve existing AI queries.

2. Design (Create Datasets)

• create-dataset: Create a new dataset.
• get-datasets: List datasets in a workspace.
• get-dataset-output: Retrieve dataset records.
• get-dataset-targetfields: Retrieve the target fields of a dataset.

3. Manage (Upload Data)

• upload-dataset-rows: Add new data rows to an existing dataset.

📦 Installation

Quick Installation

Use Smithery CLI for quick installation:

npm install -g @verodat/mcp-server

Manual Installation

1. Clone the repository:

   git clone https://github.com/Verodat/verodat-mcp-server.git
   cd verodat-mcp-server
   

2. Install dependencies:

   npm install
   

3. Build the project:

   npm run build
   

📚 Documentation

Configuration Guide

Environment Variables

• VERODAT_AI_API_KEY: Verodat API key for authentication.
• VERODAT_API_BASE_URL: The base URL of the Verodat API, defaulting to "https://verodat.io/api/v3".

Usage Guide

Available Commands

// Account and Workspace Management
get-accounts        # List accessible accounts
get-workspaces      # List workspaces within an account
get-queries         # Retrieve existing AI queries

// Dataset Operations
create-dataset      # Create a new dataset
get-datasets        # List datasets in a workspace
get-dataset-output  # Retrieve dataset records
get-dataset-targetfields # Retrieve the target fields of a dataset
upload-dataset-rows # Add new data rows to an existing dataset

// AI Operations
get-ai-context      # Get the AI context of a workspace
execute-ai-query    # Execute an AI query on a dataset

Selecting the Right Tool Category

• Read-only operations: Use the consume.js configuration.
• Create datasets: Use the design.js configuration.
• Upload data: Use the manage.js configuration.

Security Precautions

• All requests must be authenticated via an API key.
• Requests are format-validated to ensure data correctness.

Development Guide

Code Structure

• Tool handlers: Implement the functionality of each tool.
• Transport layer: Responsible for communication with AI models.
• Validation: Use Zod schemas to ensure correct data formats.

Debugging Tools

Run the following command to start the debugging tool:

npm run inspector

This will provide a URL to access the debugging tool.

📄 Support

• Documentation: Verodat Documentation
• Issues: GitHub Issues
• Community: Verodat Community

With the above information, you can comprehensively understand and use the Verodat MCP Server. If you have any questions or suggestions, please feel free to contact the support team.