verodat

Verodat MCP Server exposes three tool categories—Consume, Design, and Manage—that AI models can access via configured MCP endpoints. Consume provides read-only data retrieval capabilities such as listing accounts, workspaces, datasets, and AI-context queries. Design adds the ability to create datasets and extend the data schema, while Manage includes data upload capabilities to existing datasets. To use these tools with an AI model or interface (for example Claude Desktop), configure an MCP server entry pointing to the appropriate built JS file (consume.js, design.js, or manage.js). Each server variant expects Verodat API authentication via an API key and a base URL to Verodat’s API. Through the configuration, you can selectively enable one or more categories depending on your needs, allowing you to perform read operations, create datasets, or upload data as part of your workflow. When integrated with Claude Desktop, you can select the corresponding server entry (consume, design, or manage) to access the specific toolsets and perform actions such as retrieving datasets, creating new datasets, or uploading rows to datasets.

Prerequisites:

• Node.js (v18 or higher)
• Git
• Claude Desktop (for Claude integration)
• Verodat account and AI API key

Quick Start (Smithery):

1. Install Verodat MCP Server for Claude Desktop via Smithery:

npx -y @smithery/cli install @Verodat/verodat-mcp-server --client claude

Manual Installation:

1. Clone the repository:

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

2. Install dependencies and build:

npm install
npm run build

3. Configure Claude Desktop by creating or modifying the config file:

• MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%/Claude/claude_desktop_config.json

Add the server configuration entries (consume, design, manage) as shown in the provided configuration examples, pointing to the appropriate JS file paths generated by the build (e.g., path/to/verodat-mcp-server/build/src/consume.js).

Getting Started with Verodat:

1. Sign up for a Verodat account at verodat.com
2. Generate an AI API key from your Verodat dashboard
3. Add the API key to your Claude Desktop configuration

Note: You can configure one or more MCP servers to be active in Claude Desktop by adding corresponding entries under the mcpServers section of the Claude config.

Environment variables:

• VERODAT_AI_API_KEY: Your Verodat API key for authentication
• VERODAT_API_BASE_URL: The base URL for the Verodat API (defaults to https://verodat.io/api/v3 if not specified)

Security:

• Ensure API keys are kept secret and not checked into source control
• Use proper access control for Claude Desktop configurations

Debugging:

• The MCP server communicates over stdio; use the MCP Inspector tool to debug by running npm run inspector which provides a URL to debugging utilities.

If you enable multiple categories (consume, design, manage) in Claude Desktop, ensure you supply each corresponding environment and file path as separate server entries, and only expose the tools you intend to use in production.