Teradata MCP Server

Community

Connect AI agents directly to Teradata with enterprise security and extensibility.

How to install ↓ Compare GitHub source repository ↗

Category: Database #47 of 58
Stars: ★ 58 Popular
Transport: stdio (local process)
Runtime: Python · Docker · Hosted (no local runtime)
Credentials: API key / credential required
License: MIT
Last commit: 7 days ago
Tools: 9

47FMRS · D

The server provides comprehensive MCP integration for Teradata databases, covering querying, admin, semantic layers, and RAG. Installation is straightforward and supports multiple clients. As a community project, security and support should be self-assessed.

Strongest · Setup experience 12/20 Weakest · Reliability 8/20

Reliability

8/20

Security and permissions

10/20

Maintenance

8/20

Documentation

9/20

Setup experience

12/20

Why each score

Reliability 8/20: The project has no cached manifest; the README describes a startup method (uvx) but lacks verifiable CI workflows and tests covering key paths. Tools are scattered across multiple modules, making it impossible to confirm actual behavior from static analysis. Scoring is based on README and source structure, but due to lack of execution evidence, reliability is capped at 12.
Security and permissions 10/20: The README installation example uses placeholders (<USERNAME>:<PASSWORD>@<HOST_URL>) with no real credentials. There are row limit protections (DEFAULT_ROW_LIMIT, MAX_ROW_LIMIT) and VX views security features. However, there is no detailed least-privilege or confirmation mechanism for dangerous operations. Thus score is 10.
Maintenance 8/20: The project is active (53 stars, 20 open issues), has an MIT license, and the latest release emphasizes FastMCP v3 and hooks. But there is no clear release versioning or security response channel. Contributions and developer guides are present, but governance details are incomplete, so maintenance is scored 8.
Documentation 9/20: The README provides quick-start guides for multiple clients with configuration examples and links to detailed docs. However, troubleshooting, cost details, and tool parameter limits are missing. Documentation is rich but has hidden assumptions, hence score 9.
Setup experience 12/20: The README provides clear CLI and client configuration examples, but depends on uv and a Teradata database. No Docker or cloud deployment steps are provided. Setup is clear but requires manual configuration and external dependencies.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessConnects to a database

Best for

Developers building AI agents that need direct Teradata access
Users wanting to quickly create semantic layers
Users needing database admin capabilities
Users wanting to use RAG pipelines

Not for

Non-Teradata database users
Users needing a full GUI management interface
Enterprises requiring official support (community project)

Required permissions

Requires Teradata database credentials
May need access to database objects (tables, views, stored procedures)
May execute SQL queries
May require file system access (depending on tool)

Risks and side effects

Database credentials are stored in plaintext in environment
Executing SQL queries may modify data or impact performance
Row limit protection reduces token overflow risk
Community project, security should be self-assessed

Setup

Before you start

Runtime：Python · Docker · Hosted (no local runtime)

DATABASE_URI requiredsecret Teradata database connection string in the form teradata://<USERNAME>:<PASSWORD>@<HOST_URL>:1025/<USERNAME>; use your Teradata credentials (a free ClearScape Analytics sandbox works).

Ensure you have a Teradata database (or free sandbox) and uv installed.
Add the above JSON configuration to your client's configuration.
For Claude Desktop, edit claude_desktop_config.json and add the mcpServers configuration.

claude_desktop_config.json

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

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

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

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add teradata -e 'DATABASE_URI=teradata://<USERNAME>:<PASSWORD>@<HOST_URL>:1025/<USERNAME>' -- uvx teradata-mcp-server

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After adding the config to claude_desktop_config. and restarting, check that the teradata server's tools (e.g. from the base module) appear in the client's tool list, then ask "what tables are in my database" to confirm the connection works.

Troubleshooting

Check that the DATABASE_URI environment variable is set correctly
Ensure Teradata database is reachable and credentials are valid
Check that command and args in client configuration are correct
Review logs for detailed error information

Things to try

Once connected, you can ask your AI assistant things like:

List the tables in my Teradata database
Profile the data in my sales table and plot the patterns
Explain the query plan for my last query
Check the current database capacity usage

Tools 9

base read-only

Base tools for exploring tables, profiling data, explaining results, and visualizing patterns.

dba read-only

Database admin tools for managing security, monitoring capacity, and automating backups.

qlty read-only

Quality tools for data quality analysis.

plot read-only

Plotting tools for visualizing patterns.

rag writes

RAG tools for semantic search and retrieval-augmented generation.

tdvs writes

Teradata vector store tools for vector storage.

fs writes

File system tools, possibly for file operations.

sec writes

Security tools for managing security.

Show 1 more tools

bar writes

Backup and recovery tools for automated backups.

Use cases

Exploratory data analysis

Data engineering

Agent development

Visual workflow building

Testing new local LLMs

Supported clients

Claude Desktop

VS Code

Open WebUI

Flowise

Listed from the project's documentation, not tested by this site.

Overview

Teradata MCP Server is a community-developed MCP server that enables AI agents to interact directly with Teradata databases. It provides multiple tool sets for querying and analysis, semantic layer generation, AI and RAG pipelines, and database administration. The server runs via uvx over stdio and supports various clients including Claude Desktop, VS Code + Copilot, Open WebUI, and Flowise.

Similar servers

MCP ClickHouse 79 · B

Connect ClickHouse to your AI assistants

★ 889 · Tools 4 Compare with this →

ChEMBL MCP Server 75 · B

Unofficial MCP server providing advanced access to the ChEMBL chemical database

★ 89 · Tools 27 Compare with this →

MCP Toolbox for Databases 74 · B

Google's official database MCP toolbox — define AI-accessible database tools via config

★ 16.7k Compare with this →

DBHub 74 · B

Bytebase's official, minimalist two-tool database MCP server

★ 3.7k · Tools 2 Compare with this →

Compare with other MCP servers

Source revision 3d08f8aa0744 Data synced 2026-10-19 Read the FMRS scoring method