Agent skill · Audited from source

configuring-dbt-mcp-server

skill · dbt-labs/dbt-agent-skillsApache-2.0 licenceAudited 23 Oct 2026 · source 462d9a0d3c07

★ 745pushed today
✓SAFEGrade B · 89/100 · 23 Oct 2026 · 462d9a0
745repo stars · whole dbt-labs/dbt-agent-skills

Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 462d9a0d3c07 and found no filesystem, network or shell access and no hard-coded secrets. What it does: Generates MCP server configuration JSON, resolves authentication setup, and validates server connectivity for dbt. Use when setting up, configuring, or troubleshooting the dbt MCP server for AI tools like Claude…

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “configuring-dbt-mcp-server (dbt-labs/dbt-agent-skills)”. Retrieved 23 Oct 2026. https://modelsatlas.com/skills/dbt-labs/configuring-dbt-mcp-server

Trust audit

Audit 0.4.1 · source read 23 Oct 2026

Trust score89/100✓SAFE B
0 critical0 high0 medium0 low1 info

SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 13,985 of 14,816 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS1
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (1)

INFOPrompt injectionprompt.authority_framing · SKILL.md:59
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
| dbt Core/Fusion users | dbt Cloud without local project | Full access to both |
Why
No further detail recorded for this rule.
Fix
—

Audit 0.4.1 · source sha 462d9a0d3c07 · 23 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisherdbt-labs GitHub organizationOwner of the repository this skill is read from
  • Source folderdbt-labs/dbt-agent-skills/skills/dbt/skills/configuring-dbt-mcp-serverThe exact folder we audited
  • LicenceApache-2.0As the repository states it
  • Commit pinned for this verdict462d9a0d3c07A new audit runs when the source changes
  • Repository statusNot archivedPushed today
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Generates MCP server configuration JSON, resolves authentication setup, and validates server connectivity for dbt. Use when setting up, configuring, or troubleshooting the dbt MCP server for AI tools like Claude…

---
name: configuring-dbt-mcp-server
description: Generates MCP server configuration JSON, resolves authentication setup, and validates server connectivity for dbt. Use when setting up, configuring, or troubleshooting the dbt MCP server for AI tools like Claude Desktop, Claude Code, Cursor, or VS Code.
user-invocable: false
metadata:
  author: dbt-labs
---

# Configure dbt MCP Server

## Overview

The dbt MCP server connects AI tools to dbt's CLI, Semantic Layer, Discovery API, and Admin API. This skill guides users through setup with the correct configuration for their use case.

## Decision Flow

```mermaid
flowchart TB
    start([User wants dbt MCP]) --> q1{Local or Remote?}
    q1 -->|dev workflows,<br>CLI access needed| local[Local Server<br>uvx dbt-mcp]
    q1 -->|consumption only,<br>no local install| remote[Remote Server<br>HTTP endpoint]
    local --> q2{Which client?}
    remote --> q2
    q2 --> claude_desktop[Claude Desktop]
    q2 --> claude_code[Claude Code]
    q2 --> cursor[Cursor]
    q2 --> vscode[VS Code]
    claude_desktop --> config[Generate config<br>+ test setup]
    claude_code --> config
    cursor --> config
    vscode --> config
```

## Questions to Ask

### 1. Server Type
**Ask:** "Do you want to use the **local** or **remote** dbt MCP server?"

| Local Server                                                                                                                                             | Remote Server                                               |
| -------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------- |
| Runs on your machine via `uvx`                                                                                                                           | Connects via HTTP to dbt platform                           |
| Required for development (authoring models, tests, docs) but can also connect to the dbt platform for consumption (querying metrics, exploring metadata) | Best for consumption (querying metrics, exploring metadata) |
| Supports dbt CLI commands (run, build, test, show)                                                                                                       | No CLI commands (run, build, test)                          |
| Works without a dbt platform account but can also connect to the dbt platform for development (authoring models, tests, docs)                            | Requires dbt platform account                               |
| No credit consumption                                                                                                                                    | Consumes dbt Copilot credits                                |

### 2. MCP Client
**Ask:** "Which MCP client are you using?"
- Claude Desktop
- Claude Code (CLI)
- Cursor
- VS Code

### 3. Use Case (Local Server Only)
**Ask:** "What's your use case?"

| CLI Only | Platform Only | Platform + CLI |
|----------|---------------|----------------|
| dbt Core/Fusion users | dbt Cloud without local project | Full access to both |
| No platform account needed | OAuth or token auth | Requires paths + credentials |

### 4. Tools to Enable
**Ask:** "Which tools do you want enabled?" (show defaults)

| Tool Category | Default | Environment Variable |
|---------------|---------|---------------------|
| dbt CLI (run, build, test, compile) | Enabled | `DISABLE_DBT_CLI=true` to disable |
| Semantic Layer (metrics, dimensions) | Enabled | `DISABLE_SEMANTIC_LAYER=true` to disable |
| Discovery API (models, lineage) | Enabled | `DISABLE_DISCOVERY=true` to disable |
| Admin API (jobs, runs) | Enabled | `DISABLE_ADMIN_API=true` to disable |
| SQL (text_to_sql, execute_sql) | **Disabled** | `DISABLE_SQL=false` to enable |
| Codegen (generate models/sources) | **Disabled** | `DISABLE_DBT_CODEGEN=false` to enable |

## Prerequisites

### Local Server
1. **Install `uv`**: https://docs.astral.sh/uv/getting-started/installation/
2. **Have a dbt project** (for CLI commands)
3. **Find paths:**
   - `DBT_PROJECT_DIR`: Folder containing `dbt_project.yml`
     - macOS/Linux: `pwd` from project folder
     - Windows: Full path with forward slashes (e.g., `C:/Users/name/project`)
   - `DBT_PATH`: Path to dbt executable
     - macOS/Linux: `which dbt`
     - Windows: `where dbt`

### Remote Server
1. **dbt Cloud account** with AI features enabled
2. **Production environment ID** (from Orchestration page)
3. **Personal access token** or service token

See [How to Find Your Credentials](references/finding-credentials.md) for detailed guidance on obtaining tokens and IDs.

## Credential Security

- Always use environment variable references (e.g., `${DBT_TOKEN}`) instead of literal token values in configuration files that may be committed to version control
- Never log, display, or echo token values in terminal output
- When using `.env` files, ensure they are added to `.gitignore` to prevent accidental commits
- Recommend users rotate tokens regularly and use the minimum required permission set

## Configuration Templates

### Local Server - CLI Only

```json
{
  "mcpServers": {
    "dbt": {
      "command": "uvx",
      "args": ["dbt-mcp"],
      "env": {
        "DBT_PROJECT_DIR": "/path/to/your/dbt/project",
        "DBT_PATH": "/path/to/dbt"
      }
    }
  }
}
```

### Local Server - Platform + CLI (OAuth)

```json
{
  "mcpServers": {
    "dbt": {
      "command": "uvx",
      "args": ["dbt-mcp"],
      "env": {
        "DBT_HOST": "https://your-subdomain.us1.dbt.com",
        "DBT_PROJECT_DIR": "/path/to/project",
        "DBT_PATH": "/path/to/dbt"
      }
    }
  }
}
```

### Local Server - Platform + CLI (Token Auth)

```json
{
  "mcpServers": {
    "dbt": {
      "command": "uvx",
      "args": ["dbt-mcp"],
      "env": {
        "DBT_HOST": "cloud.getdbt.com",
        "DBT_TOKEN": "${DBT_TOKEN}",
        

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit 462d9a0

Download folder ↗
claude mcp add dbt -s user -- uvx dbt-mcp
claude mcp add dbt -s project -- uvx dbt-mcp
uvx --env-file .env dbt-mcp
uvx dbt-mcp

Commands as the repository documents them: shown, never run by us. Our verdict covers commit 462d9a0d3c07 only.

Works in which agents?

What its own text says, not a test

AgentStatusMore
Claude CodeMentioned in its text, not testedSkills for Claude Code · About Claude Code
Claude DesktopMentioned in its text, not testedSkills for Claude Desktop · About Claude Desktop
GitHub CopilotMentioned in its text, not testedSkills for GitHub Copilot · About GitHub Copilot
CursorMentioned in its text, not testedSkills for Cursor · About Cursor
“Mentioned” means the skill’s text names the agent. We have not run it there.

Audit history

Every audit, every source change

  1. ✓SAFE B 89 / 100

    first audit · source 462d9a0d3c07

README badge

For maintainers · shows the latest audit

ModelsAtlas: SAFE BMODELSATLASSAFE · B
[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/dbt-labs/configuring-dbt-mcp-server/badge.svg)](https://modelsatlas.com/skills/dbt-labs/configuring-dbt-mcp-server/)

The badge shows our verdict and grade for this skill, and links back to this page. It changes when the audit does.

Similar skills

Add this badge to your README

[![ModelsAtlas: SAFE B](https://modelsatlas.com/skills/dbt-labs/configuring-dbt-mcp-server/badge.svg)](https://modelsatlas.com/skills/dbt-labs/configuring-dbt-mcp-server/)

Questions about configuring-dbt-mcp-server

What does the configuring-dbt-mcp-server skill do?
A curated collection of Agent Skills for working with dbt, to help AI agents understand and execute dbt workflows more effectively.
Is configuring-dbt-mcp-server safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can configuring-dbt-mcp-server access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does configuring-dbt-mcp-server work with?
Its documentation mentions claude-code, claude-desktop, copilot and cursor. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (462d9a0d3c07), read on 2026-10-23. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.