AUTOMATIONSWITCH

dbt MCP

by dbt Labs

Official dbt Labs MCP server covering SQL execution, Semantic Layer queries, Discovery API, dbt CLI, Admin API, dbt Codegen, dbt LSP, and Product Docs. 41 commits on main in the last 30 days, joint top cadence in this batch.

551·9 tools·Released AUG 2025·Apache-2.0
uvx dbt-mcp
Share:

Official dbt Labs MCP server covering SQL execution, Semantic Layer queries, Discovery API, dbt CLI, Admin API, dbt Codegen, dbt LSP, and Product Docs. 41 commits on main in the last 30 days, joint top cadence in this batch with Azure DevOps and Canvas LMS.

Reviewed by M. Nouriel · MAY 2026

INSTALL THIS SERVER

Requires authenticationDBT_CLOUD_API_TOKEN (for Admin API and Discovery), DBT_CLOUD_ACCOUNT_ID. CLI tools use the local dbt project profiles for warehouse credentials.
{ "mcpServers": { "dbt": { "command": "uvx", "args": [ "dbt-mcp" ], "env": { "DBT_CLOUD_API_TOKEN": "<your-dbt-cloud-token>", "DBT_CLOUD_ACCOUNT_ID": "<your-account-id>", "DBT_PROJECT_DIR": "/path/to/dbt/project" } } } }
PrereqRequires uv installed. Set DBT_CLOUD_API_TOKEN and DBT_CLOUD_ACCOUNT_ID for Admin API and Discovery. Set DBT_PROJECT_DIR for local CLI tools. PyPI: `dbt-mcp`. Path: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS).
{ "mcpServers": { "dbt": { "command": "uvx", "args": [ "dbt-mcp" ], "env": { "DBT_CLOUD_API_TOKEN": "<your-dbt-cloud-token>", "DBT_CLOUD_ACCOUNT_ID": "<your-account-id>", "DBT_PROJECT_DIR": "/path/to/dbt/project" } } } }
{ "mcpServers": { "dbt": { "command": "uvx", "args": [ "dbt-mcp" ], "env": { "DBT_CLOUD_API_TOKEN": "<your-dbt-cloud-token>", "DBT_CLOUD_ACCOUNT_ID": "<your-account-id>", "DBT_PROJECT_DIR": "/path/to/dbt/project" } } } }
{ "mcpServers": { "dbt": { "command": "uvx", "args": [ "dbt-mcp" ], "env": { "DBT_CLOUD_API_TOKEN": "<your-dbt-cloud-token>", "DBT_CLOUD_ACCOUNT_ID": "<your-account-id>", "DBT_PROJECT_DIR": "/path/to/dbt/project" } } } }
{ "mcpServers": { "dbt": { "command": "uvx", "args": [ "dbt-mcp" ], "env": { "DBT_CLOUD_API_TOKEN": "<your-dbt-cloud-token>", "DBT_CLOUD_ACCOUNT_ID": "<your-account-id>", "DBT_PROJECT_DIR": "/path/to/dbt/project" } } } }

9 TOOLS AVAILABLE

execute_sql
Execute SQL against the dbt warehouse connection
Read
query_semantic_layer
Run governed metric queries via the dbt Semantic Layer
Read
get_models
List dbt models with metadata via the Discovery API
Read
get_model_lineage
Retrieve upstream and downstream lineage for a model
Read
dbt_run
Run dbt models via the CLI
Write
dbt_test
Run dbt tests via the CLI
Read

OUR ASSESSMENT

Strengths
  • Official dbt Labs maintenance.
  • 551 GitHub stars and Apache-2.0 licence.
  • 41 commits on main in the last 30 days.
  • Eight tool families cover the full dbt workflow: SQL, Semantic Layer, Discovery, CLI, Admin API, Codegen, LSP, Docs.
  • Experimental MCP Bundle support flagged in the README for early-access features.
  • Architecture documented in the README: clear separation between local dbt project tools and dbt Cloud Admin API.
Weaknesses
  • Setup complexity: requires a dbt project on disk plus optional dbt Cloud credentials for Admin API and Discovery surfaces.
  • Semantic Layer and Discovery require dbt Cloud subscription; CLI and SQL work against any dbt project.
  • The MCP Bundle feature is marked experimental; behaviour may evolve.
Security Notes

The MCP carries dbt Cloud API tokens (when configured) plus warehouse credentials from the dbt project profile. Scope dbt Cloud tokens to read-only when the agent does diagnostic work; the Admin API token grants full job management. The warehouse connection uses the credentials in the dbt project profile; for production agents, run the MCP under a dedicated read-only warehouse role.

Best For

Data teams using dbt for analytics engineering who want agents to execute SQL, query the Semantic Layer, inspect model lineage, and trigger dbt Cloud jobs from inside the editor; analytics engineers who want CLI run/test/build flows accessible from agent prompts; teams scaffolding new dbt models via Codegen.

TECHNICAL DETAILS

Language
python
Transport
stdio
Clients
Claude DesktopClaude CodeCursorVS CodeWindsurf
License
Apache-2.0
GitHub
dbt-labs/dbt-mcp · ★ 551
npm
dbt-mcp
Last Release
dbt-mcp (PyPI latest)MAY 3, 2026
First Released
AUG 1, 2025

ADOPTION METRICS

// GitHub Stars
551

// Reading this551 stars on the dbt-labs/dbt-mcp repo. 41 commits on main in the last 30 days, joint top cadence in this batch.

// Popularity Rank
#1
Globally · #4 in Data / Analytics

// Reading thisHighest-quality data MCP at time of review for analytics-engineering workflows.

SOURCES & VERIFICATION

We don't take any single directory's word for it. Before scoring, we cross-reference 5 public MCP sources, install the server ourselves against the clients we cover, and record when we last re-verified.

01
Discovered
Manual submission
First indexed MAY 3, 2026
02
Cross-referenced
5 directories
PulseMCP, MCP.so, Glama, Smithery, Official MCP Registry
03
Verified against
Claude Desktop, Cursor
Installed and tested across clients
04
Last re-checked
MAY 3, 2026
Weekly re-verification
// How other directories see it

The same server, 5 different lenses. We reconcile these signals into our editorial score, which is why our number sometimes diverges from a directory-aggregate star count.

SourceTheir ratingTheir star countTheir downloadsLast synced
AutomationSwitch This page4.5editorial551MAY 3, 2026
PulseMCP— unratedunavailableunavailableMAY 3, 2026
MCP.so— unratedunavailableunavailableMAY 3, 2026
Glama— unratedunavailableunavailableMAY 3, 2026
Smithery— unratedunavailableunavailableMAY 3, 2026
Official MCP Registry— unratedunavailableunavailableMAY 3, 2026

// Counts are directory-reported; we don't adjust them. Discrepancies usually come from different snapshot times or star-caching.

// Get in touch

DISCUSS YOUR
MCP REQUIREMENTS.

Evaluating a server, scoping an internal deployment, or working out whether MCP is the right fit at all. Start the conversation and we will point you at the right piece of the ecosystem.

Discuss Your MCP Requirements →