Teradata MCP Server for Claude
Connect Claude to Teradata — list databases and tables, inspect DDL, run SQL queries, preview data, analyze column quality, and explore DBA diagnostics — with the official Teradata MCP server supporting optional ML tool expansion via the teradataml Python package.
Open the source and read safety notes before installing.
Safety notes
- The `base_query` tool executes arbitrary SQL on your Teradata system — only connect with credentials scoped to the minimum required permissions.
- Row limits (`DEFAULT_ROW_LIMIT`, `MAX_ROW_LIMIT`) prevent accidental full-table scans returning millions of rows; leave defaults in place unless you need larger result sets.
Privacy notes
- Query results, schema definitions, DDL, and session/system metadata from your Teradata instance are surfaced in Claude's context.
- Your database credentials are embedded in the `DATABASE_URI` — keep it in the MCP config env and never commit it to version control.
Prerequisites
- A Teradata database instance accessible from your network.
- A Teradata username and password with appropriate query permissions.
- Python with `uvx` available.
- An MCP client such as Claude Code or Claude Desktop.
Schema details
- Install type
- cli
- Troubleshooting
- No
- Scope
- Source repo
- Estimated setup
- 10 minutes
- Difficulty
- intermediate
- Website
- https://www.teradata.com
Full copyable content
{
"mcpServers": {
"teradata": {
"command": "uvx",
"args": ["teradata-mcp-server"],
"env": {
"DATABASE_URI": "teradata://username:password@host:1025/databasename"
}
}
}
}About this resource
Overview
The Teradata MCP Server is the official Model Context Protocol server from Teradata,
giving Claude structured access to Teradata databases. It covers database and table discovery,
DDL inspection, SQL execution, data previewing, DBA diagnostics, column quality analysis, and
session monitoring. Install the optional teradataml Python package to unlock ~89 additional
ML tools for statistical modeling and machine learning workflows. Licensed under MIT.
Key capabilities
- Database discovery — list all databases and their tables/views.
- Schema inspection — retrieve full DDL for any table or view.
- SQL execution — run arbitrary SQL queries with configurable row limits.
- Data preview — preview rows from tables quickly without writing SQL.
- Column quality — statistical summaries and missing value analysis.
- DBA diagnostics — session info, system space, and user/permission queries.
- ML tools (optional) —
teradatamladds ~89 dynamic tools: KMeans, XGBoost, AutoML, etc.
Tools (key selection)
| Tool | Category | Purpose |
|---|---|---|
base_databaseList |
Discovery | List all databases |
base_tableList |
Discovery | List tables and views in a database |
base_tableDDL |
Schema | Show DDL for a table |
base_query |
SQL | Execute arbitrary SQL |
base_tablePreview |
Data | Preview rows from a table |
dba_sessionInfo |
DBA | Active session information |
dba_systemSpace |
DBA | System-wide space summary |
sec_userDbPermissions |
Security | User database permissions |
qlty_columnSummary |
Quality | Column-level statistics |
qlty_missingValues |
Quality | Missing value analysis |
Configuration options
| Env var | Default | Purpose |
|---|---|---|
DATABASE_URI |
— | Required. Full connection URI |
LOGMECH |
TD2 |
Auth mechanism: TD2 or LDAP |
TD_POOL_SIZE |
5 |
Connection pool size |
DEFAULT_ROW_LIMIT |
1000 |
Default row cap for results |
MAX_ROW_LIMIT |
50000 |
Hard ceiling on results |
MCP_TRANSPORT |
stdio |
Transport: stdio, sse, or streamable-http |
How it compares
| Server | Teradata | SQL execution | DDL inspection | Column quality | ML tools |
|---|---|---|---|---|---|
| Teradata MCP | Yes | Yes | Yes | Yes | Opt-in (teradataml) |
| Snowflake MCP | No | Yes | Yes | No | No |
| BigQuery MCP | No | Yes | Yes | No | No |
| DuckDB MCP | No | Yes | Yes | No | No |
Teradata MCP is the only data warehouse MCP with built-in column quality analysis and optional
ML model tools via teradataml.
Installation
Claude Code
claude mcp add teradata \
-e DATABASE_URI="teradata://username:password@host:1025/databasename" \
-- uvx teradata-mcp-server
Claude Desktop
{
"mcpServers": {
"teradata": {
"command": "uvx",
"args": ["teradata-mcp-server"],
"env": {
"DATABASE_URI": "teradata://username:password@host:1025/databasename"
}
}
}
}
For LDAP authentication, add "LOGMECH": "LDAP" to the env block.
Requirements
- Teradata database instance (on-premises or Teradata VantageCloud).
- Python with
uvx(fromuv). - An MCP client (Claude Code or Claude Desktop).
- Optional:
pip install teradatamlfor ML tools.
Security
- Use credentials scoped to least privilege — read-only if SQL execution isn't needed.
base_queryruns arbitrary SQL; ensure the Teradata user does not have destructive permissions.- Keep the
DATABASE_URIsecret; it contains your username and password.
Source Verification Notes
Verified on 2026-06-18:
- Official repository
Teradata/teradata-mcp-server(MIT) on PyPI asteradata-mcp-serverdocuments theDATABASE_URIconnection URI format, all tool categories (base, dba, sec, qlty), optional teradataml ML tools, connection pool and row limit configuration, and multiple transport modes. - Claude Code MCP documentation at
code.claude.com/docs/en/mcpdescribes the stdio connector pattern used above.
Source citations
Add this badge to your README
How it compares
Teradata MCP Server for Claude side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
| Field | Teradata MCP Server for Claude Connect Claude to Teradata — list databases and tables, inspect DDL, run SQL queries, preview data, analyze column quality, and explore DBA diagnostics — with the official Teradata MCP server supporting optional ML tool expansion via the teradataml Python package. Open dossier | MCP SQLite Server for Claude Read and write any local SQLite database from Claude — get database info, list tables, inspect schemas, create/read/update/delete records, and run raw SQL queries — with the MCP SQLite server that connects to any `.db` file you specify. Open dossier | Metabase MCP Server for Claude Query and explore your Metabase instance from Claude — list dashboards and cards, retrieve and execute saved questions, run SQL queries, export large result sets, and search across all Metabase resources — with the Metabase MCP server optimized for 90% token reduction. Open dossier | Appwrite MCP Server for Claude Connect Claude to the full Appwrite backend platform — databases, authentication, serverless functions, teams, messaging, and storage — with the official Appwrite MCP server using dynamic dispatch across the entire Appwrite API surface. Open dossier |
|---|---|---|---|---|
| Trust | ||||
| Install risk | Review first | Review first | Review first | Review first |
| Notes | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ | Safety ✓ Privacy ✓ |
| Category | mcp | mcp | mcp | mcp |
| Source | source-backed | source-backed | source-backed | source-backed |
| Author | Teradata | jparkerweb | Jericho Sequitin | Appwrite |
| Added | 2026-06-18 | 2026-06-18 | 2026-06-18 | 2026-06-18 |
| Platforms | Claude CodeCodexCursorClaude Desktop | Claude CodeClaude Desktop | Claude CodeClaude Desktop | Claude CodeClaude Desktop |
| Source repo | — | — | — | — |
| Safety notes | ✓The `base_query` tool executes arbitrary SQL on your Teradata system — only connect with credentials scoped to the minimum required permissions. Row limits (`DEFAULT_ROW_LIMIT`, `MAX_ROW_LIMIT`) prevent accidental full-table scans returning millions of rows; leave defaults in place unless you need larger result sets. | ✓The `query` tool executes arbitrary SQL including INSERT, UPDATE, and DELETE statements — it is not read-only. Point the server at a copy or test database to avoid accidental data loss before trusting it with production data. | ✓Read-only mode is enabled by default (`METABASE_READ_ONLY_MODE=true`), restricting Claude to SELECT queries only. The `execute` tool runs SQL against your connected databases — ensure the Metabase service account has appropriate read permissions. | ✓Mutation operations (write, update, delete) require `confirm_write=true` to be passed explicitly, adding a confirmation step before any destructive action. The API key scope determines what operations are available — use least-privilege scopes for your use case. |
| Privacy notes | ✓Query results, schema definitions, DDL, and session/system metadata from your Teradata instance are surfaced in Claude's context. Your database credentials are embedded in the `DATABASE_URI` — keep it in the MCP config env and never commit it to version control. | ✓All database content including table schemas, record data, and query results are surfaced in Claude's context — do not connect databases containing sensitive personal or credential data unless you intend to share it with the model. No credentials are required and no data is sent to external services — the server operates entirely locally. | ✓Dashboard data, card results, database schema information, and query results from your Metabase instance are surfaced in Claude's context. Your `METABASE_API_KEY` (or email/password) grants Metabase API access — keep credentials in the MCP config env. | ✓User records, database documents, function logs, team memberships, and messaging data from your Appwrite project are surfaced in Claude's context. Your `APPWRITE_API_KEY` grants project-level access — keep it in the MCP config env and never commit it to version control. |
| Prerequisites |
|
|
|
|
| Install | | | | |
| Config | | | | |
| Citations | ||||
| Claim | Unclaimed | Unclaimed | Unclaimed | Unclaimed |
Featured in
Signals
Loading live community signals…
A short, calm digest of reviewed Claude resources. Unsubscribe any time.