Teradata MCP Server Template

Name: Teradata MCP Server Template
Author: Teradata

@Teradata

代码仓库

关于 Teradata MCP Server Template

The community development of a MCP server for a Teradata database

基本信息

分类

开发工具

许可证

MIT

运行时

python

传输方式

stdio

发布者

Teradata

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "teradata": {
      "command": "uvx",
      "args": [
        "teradata-mcp-server"
      ],
      "env": {
        "DATABASE_URI": "teradata://<USERNAME>:<PASSWORD>@<HOST_URL>:1025/<USERNAME>"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Teradata MCP Server?

Teradata MCP Server connects AI agents directly to Teradata databases with enterprise security and extensibility. It enables natural language querying, semantic layers, AI and RAG pipelines, and database administration. This server is for data engineers, analysts, and developers who want to interact with Teradata using AI assistants.

How to use Teradata MCP Server?

Install via uvx and configure with a DATABASE_URI environment variable in your client’s MCP settings (e.g., Claude Desktop claude_desktop_config.json). No local installation is required—just add the JSON entry with "command": "uvx" and "args": ["teradata-mcp-server"]. You can also run it as an HTTP server, Docker container, or cloud service.

Key features of Teradata MCP Server

Query and analyze tables without writing SQL
Create custom semantic layers from YAML files
Semantic search and retrieval-augmented generation
Database administration including security and backup tools
Configurable row limit protection against token overflow
Hooks capability for custom monitoring and audit trails

Use cases of Teradata MCP Server

Exploratory data analysis via natural language chat (Claude Desktop)
Data engineering and agent development with VS Code + Copilot
Testing new LLMs locally using Open WebUI
Building custom client applications with provided code examples
Visual agent building with Flowise drag-and-drop workflows

FAQ from Teradata MCP Server

What is Teradata MCP Server?

It is an MCP server that connects AI agents to Teradata databases, providing tools for querying, semantic layers, AI/RAG pipelines, and database administration.

How do I install Teradata MCP Server?

No local installation is needed—use uvx with a DATABASE_URI environment variable in your client’s MCP configuration. You can also run it via Docker, HTTP server, or as a cloud service.

What are the prerequisites?

A Teradata database (a free sandbox is available) and uv (see uv installation guide). The DATABASE_URI must be set to teradata://<USERNAME>:<PASSWORD>@<HOST_URL>:1025/<USERNAME>.

What tools does it include?

It includes tool categories: base (query and analyze), dba (database admin), qlty (data quality), plot (visualization), rag (retrieval-augmented generation), tdvs (vector storage), fs (file system), sec (security), and bar (backup/restore).

Can I extend the server?

Yes. You can add custom logic using hooks (monitoring, audit, validation), define domain-specific semantic layers in YAML, or deploy custom middleware (e.g., performance monitoring patterns).