home / mcp / teradata mcp server
Provides modular Teradata data tools and prompts to query, analyze, and manage data through MCP-enabled agents.
Configuration
View docs{
"mcpServers": {
"blitzstermayank-mcp": {
"url": "https://mcp.teradata.example.com/mcp",
"headers": {
"DATABASE_URI": "teradata://USER:PASSWORD@HOST:1025/USER"
}
}
}
}The Teradata MCP Server provides a set of modular AI-assisted tools and prompts to interact with Teradata databases. You can query, explore, and manage data, access feature stores, verify data quality, and support database administration and data science workflows through cohesive MCP-enabled agents and prompts.
You interact with the Teradata MCP Server through an MCP client (such as Claude Desktop) by connecting to the server and selecting the Teradata module set. The server exposes grouped tools and prompts that guide you to search vector stores, query the Teradata platform, access structured data models, assess data quality, perform DBA tasks, and enable data science workflows. Use these tools to build agents that can search, reason about, and manipulate Teradata data in natural language or scripted prompts.
Typical workflows include: - Searching vector stores and performing RAG-enabled retrieval to surface relevant Teradata data. - Querying base data and semantic layer definitions to navigate and model data. - Accessing the Teradata Enterprise Feature Store and running domain-specific prompts for business data models. - Running data quality checks to explore data health and identify anomalies. - Utilizing DBA and Security Tools to administer the platform and enforce security controls. - Building AI-driven pipelines with Teradata Vector Store Tools and plotting utilities for data visualization.
{
"mcpServers": {
"teradata": {
"command": "uvx",
"args": ["teradata-mcp-server"],
"env": {
"DATABASE_URI": "teradata://<USERNAME>:<PASSWORD>@<HOST_URL>:1025/<USERNAME>"
}
}
}
}Prerequisites include having Teradata credentials and a running MCP runtime. Install the MCP runtime and ensure you can reach the Teradata database URL. Use the following local configuration example to start the Teradata MCP Server through your MCP client.
If you plan to use Claude Desktop, place the above JSON under your client configuration so the MCP server is loaded as teradata with the proper DATABASE_URI. Update the placeholder fields with your actual credentials and host details.
Tools and prompts to search and manage vector stores and perform retrieval-augmented generation (RAG) applications.
Base query tools to navigate Teradata data and perform standard SQL-like queries against core tables.
Access and manage the Teradata Enterprise Feature Store for feature engineering and serving features to models.
Define and utilize semantic layer definitions for domain-specific data models and prompts.
Data quality tools and prompts to drive exploratory data analysis and data health checks.
DBA-focused prompts and resources for platform administration tasks and security operations.
Security-oriented prompts and tools to help enforce access controls and monitor risk.
Teradata Vector Store Tools for storing, retrieving, and querying embedding vectors.
Tools to work with TeradataML and related data science workflows.
Visualization helpers to plot data insights and results from MCP prompts.