
dbt Agent Skills
Skilldbt Labs 官方技能合集:编写与修改 dbt 模型、补单元测试与文档、搭建 MetricFlow 语义层、排查任务失败,以及 dbt 版本升级与跨平台迁移。多数技能需要已安装 dbt 并有现成项目。
dbt Agent Skills
A curated collection of Agent Skills for working with dbt. These skills help AI agents understand and execute dbt workflows more effectively.
What are Agent Skills?
Agent Skills are folders of instructions, scripts, and resources that agents can discover and use to do things more accurately and efficiently.
How They Work
These skills are not slash commands or user-invoked actions. Once installed, the agent automatically loads the relevant skill when your prompt matches its use case. Just describe what you need in natural language and the agent handles the rest. See skill invocation control for more details.
What's Included
- Analytics engineering: Build and modify dbt models, write tests, explore data sources
- Semantic layer: Create metrics, dimensions, and semantic models with MetricFlow
- dbt Mesh: Work with multi-project setups, cross-project refs, model governance (contracts, versions, access)
- Platform operations: Troubleshoot job failures, configure the dbt MCP server
- Migration: Move projects from dbt Core to dbt v2
Installation
Claude Code
Add the dbt skills marketplace and install the plugins:
# Add the marketplace
/plugin marketplace add dbt-labs/dbt-agent-skills
# Install the dbt skills (analytics engineering, semantic layer, testing, etc.)
/plugin install dbt@dbt-agent-marketplace
# Install the migration skills (typically a one-off — not needed for every session)
/plugin install dbt-migration@dbt-agent-marketplace
Other AI Clients
Vercel Skills CLI
Use the Vercel Skills CLI to install skills from this repository. Supports 30+ AI agents including Cursor, Cline, GitHub Copilot, and others.
# Preview available skills
npx skills add dbt-labs/dbt-agent-skills --list
# Install all skills
npx skills add dbt-labs/dbt-agent-skills
# Install only the dbt skills (analytics engineering, semantic layer, etc.)
npx skills add dbt-labs/dbt-agent-skills/skills/dbt
# Install only the migration skills
npx skills add dbt-labs/dbt-agent-skills/skills/dbt-migration
# Install a specific skill
npx skills add dbt-labs/dbt-agent-skills --skill using-dbt-for-analytics-engineering
# Install globally (available in all projects, stored in ~/.<agent>/skills/)
npx skills add dbt-labs/dbt-agent-skills --global
# Check for updates
npx skills check
# Update installed skills
npx skills update
Tessl
Install skills using Tessl, a package manager for agent skills:
# Install all skills
tessl install dbt-labs/dbt-agent-skills
# Install a specific skill
tessl install dbt-labs/dbt-agent-skills --skill using-dbt-for-analytics-engineering
# Install from GitHub directly
tessl install github:dbt-labs/dbt-agent-skills
Browse the tile on the Tessl registry.
dbt packages (dbt deps)
Install skills as dbt package dependencies. This repository publishes three independent packages, so you can install any combination:
| Package | Subdirectory | Contents |
|---|---|---|
dbt_agent_skills | skills/dbt | Everyday dbt development skills |
dbt_migration_skills | skills/dbt-migration | One-off migration and upgrade skills |
dbt_extras_skills | skills/dbt-extras | Extra skills, such as Mermaid DAG diagrams |
Add any of them to your project's packages.yml:
packages:
# Everyday dbt skills
- git: "https://github.com/dbt-labs/dbt-agent-skills.git"
subdirectory: "skills/dbt"
revision: main
# Migration skills (typically a one-off)
- git: "https://github.com/dbt-labs/dbt-agent-skills.git"
subdirectory: "skills/dbt-migration"
revision: main
# Extra skills
- git: "https://github.com/dbt-labs/dbt-agent-skills.git"
subdirectory: "skills/dbt-extras"
revision: main
Then opt in by setting ai_provider in your own project's dbt_project.yml (it is ignored if set inside a package). Without it, dbt deps succeeds but installs no skills:
flags:
ai_provider: claude # Claude Code -> .claude/skills
Supported values include claude, codex, cursor, gemini, openai, and wizard (dbt Wizard CLI). Everything except claude installs to .agents/skills. To target several, use a list (ai_provider: [wizard, claude]), or pass --ai-provider claude,codex on the command line.
dbt deps
[!TIP]
revision: maintracks the latest changes. For reproducible installs, pinrevisionto a Git tag or commit SHA instead.
See the dbt documentation on package skills for the full configuration reference, including how to disable individual skills.
Compatible Agents
These skills work with AI agents that support the Agent Skills format.
Available Skills
dbt (analytics engineering)
| Skill | Description |
|---|---|
using-dbt-for-analytics-engineering | Build and modify dbt models, debug errors, explore data sources, write tests |
adding-dbt-unit-test | Add unit tests for dbt models, practice test-driven development |
maintaining-dbt-documentation | Audit model/column doc coverage and draft missing descriptions in the project's own style |
building-dbt-semantic-layer | Create semantic models, metrics, and dimensions with MetricFlow |
answering-natural-language-questions-with-dbt | Answer business questions by querying the semantic layer |
working-with-dbt-mesh | Implement dbt Mesh governance (contracts, access, groups, versions) and cross-project collaboration |
troubleshooting-dbt-job-errors | Diagnose and resolve dbt platform job failures |
configuring-dbt-mcp-server | Set up the dbt MCP server for Claude, Cursor, or VS Code |
fetching-dbt-docs | Look up dbt documentation efficiently |
running-dbt-commands | Run dbt CLI commands with correct flags, selectors, and parameter formats |
querying-the-dbt-information-schema | Answer questions about project metadata (models, tests, columns, lineage) with SQL on dbt v2 |
dbt-migration (one-off use)
These skills are typically used once during a migration project rather than in every agent session.
| Skill | Description |
|---|---|
migrating-dbt-core-to-v2 | Migrate dbt projects from dbt Core to dbt v2 |
migrating-dbt-project-across-platforms | Migrate dbt projects across data platforms |
upgrading-dbt | Upgrade a dbt-core v1 project (1.3-1.7) to the latest dbt, applying breaking, behavior, and deprecated changes |
Prerequisites
Most skills assume:
- dbt is installed and configured
- A dbt project with
dbt_project.ymlexists - Basic familiarity with dbt concepts (models, tests, sources)
Some skills like fetching-dbt-docs and configuring-dbt-mcp-server can be used without an existing project.
Contributing
We welcome contributions! Whether you want to add a new dbt skill, improve existing ones, or fix issues, please see our Contributing Guide.
Format Specification
All skills in this repository follow the Agent Skills specification to ensure compatibility across different agent products.
Resources
Community
- Issues: Report problems or suggest new skills
- Discussions: Share use cases and patterns
- Pull Requests: Contribute new skills or improvements
- Star this repository if you find it useful!
License
See LICENSE for details.
Skill Evaluation
See evals/README.md for the A/B testing tool to compare skill variations.
相关资源

SQLMesh
CLI用 SQL 或 Python 编写数据转换模型的框架。提供类似 Terraform 的 plan / apply 流程、虚拟数据环境、单元测试和表差异比对,能在 10 多种 SQL 方言之间转译。Linux 基金会项目,可以替代 dbt。
SQLMesh/sqlmesh3.3k
dbt
CLI数仓建模与 ELT 转换的命令行框架:用 SQL 编写模型,自动处理依赖、数据测试与文档生成。v2 用 Rust 重写为单一二进制,dbt Labs 官方维护,需配置数据仓库连接。
dbt-labs/dbt14k
dbt Agent Skills
Plugindbt Labs 官方插件市场,提供 dbt 与 dbt-migration 等插件:编写模型与单元测试、构建语义层指标、dbt Mesh 治理、排查作业失败,以及版本升级与跨平台迁移。多数技能需要现成的 dbt 项目。
dbt-labs/dbt-agent-skills737
dbt MCP Server
MCP让 agent 读取 dbt 项目的模型、血缘与语义层指标,执行 build、run、test 等命令,并调用 dbt 平台的 Discovery 与 Admin API。dbt Labs 官方维护;语义层与平台类工具需要 dbt 平台账号。
dbt-labs/dbt-mcp609
Databricks Agent Skills
SkillDatabricks 官方技能库,指导 agent 在湖仓上开发 Lakeflow 数据管道与作业、Apps、Lakebase、向量搜索、模型服务和 AI/BI 看板。需要安装 Databricks CLI。
databricks/databricks-agent-skills352
SQLFluff
CLI支持二十多种 SQL 方言的 linter 与格式化工具,可自动修复大部分问题,兼容 Jinja 模板与 dbt 项目(dbt 需装插件),适合在数仓代码评审和 CI 中使用。
sqlfluff/sqlfluff10k