Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This collection provides specialized tools for analytics engineers managing complex dbt projects. It streamlines the process of mapping model lineage and resolving dialect differences during warehouse migrations.
By leveraging real-time compilation and manifest parsing, the skill set helps teams visualize dependencies through Mermaid diagrams and remediate SQL incompatibilities when moving projects between platforms like Snowflake and Databricks.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
Setup
Claude Code dbt plugin
Installs: The complete dbt analytics-engineering plugin bundle from dbt-agent-marketplace. · Claude Code
Before you start
Claude Code
Most included skills assume dbt is installed and configured, a dbt project containing dbt_project.yml, and basic familiarity with dbt concepts; fetching-dbt-docs and configuring-dbt-mcp-server can be used without an existing project.
For building the dbt Semantic Layer, the legacy specification requires dbt Core 1.6–1.11; the latest specification requires dbt Core 1.12+ or Fusion.
For local dbt MCP server setup, uv must be installed, and a dbt project is required for CLI commands. Remote setup instead requires a dbt Cloud account with AI features enabled, a production environment ID, and a personal access token or service token.
In a Claude Code agent session, add the author's marketplace.
/plugin marketplace add dbt-labs/dbt-agent-skills
In the same Claude Code agent session, install the dbt plugin bundle.
Installs: All skills from the repository using the Vercel Skills CLI, with its default non-global scope. · Cursor, Cline, GitHub Copilot
Before you start
Vercel Skills CLI invoked through npx
Most skills assume dbt is installed and configured, a dbt project containing dbt_project.yml, and basic familiarity with dbt concepts; fetching-dbt-docs and configuring-dbt-mcp-server can be used without an existing project.
For building the dbt Semantic Layer, the legacy specification requires dbt Core 1.6–1.11; the latest specification requires dbt Core 1.12+ or Fusion.
For local dbt MCP server setup, uv must be installed, and a dbt project is required for CLI commands. Remote setup instead requires a dbt Cloud account with AI features enabled, a production environment ID, and a personal access token or service token.
For cross-platform migration, dbt Fusion must be installed and working, and it must connect to both source and target platforms.
dbt v2 is required when using the included migrating-dbt-core-to-v2 skill.
Run this terminal command to install all skills from the repository with the CLI's default scope.
Installs: Only the complete skills/dbt analytics-engineering bundle. · Cursor, Cline, GitHub Copilot
Before you start
Vercel Skills CLI invoked through npx
Most included skills assume dbt is installed and configured, a dbt project containing dbt_project.yml, and basic familiarity with dbt concepts; fetching-dbt-docs and configuring-dbt-mcp-server can be used without an existing project.
For building the dbt Semantic Layer, the legacy specification requires dbt Core 1.6–1.11; the latest specification requires dbt Core 1.12+ or Fusion.
For local dbt MCP server setup, uv must be installed, and a dbt project is required for CLI commands. Remote setup instead requires a dbt Cloud account with AI features enabled, a production environment ID, and a personal access token or service token.
Run this terminal command to install only the dbt analytics-engineering bundle.
Installs: All repository skills, globally available in all projects and stored under ~/.<agent>/skills/. · Cursor, Cline, GitHub Copilot
Before you start
Vercel Skills CLI invoked through npx
Most skills assume dbt is installed and configured, a dbt project containing dbt_project.yml, and basic familiarity with dbt concepts; fetching-dbt-docs and configuring-dbt-mcp-server can be used without an existing project.
For building the dbt Semantic Layer, the legacy specification requires dbt Core 1.6–1.11; the latest specification requires dbt Core 1.12+ or Fusion.
For local dbt MCP server setup, uv must be installed, and a dbt project is required for CLI commands. Remote setup instead requires a dbt Cloud account with AI features enabled, a production environment ID, and a personal access token or service token.
For cross-platform migration, dbt Fusion must be installed and working, and it must connect to both source and target platforms.
dbt v2 is required when using the included migrating-dbt-core-to-v2 skill.
Run this terminal command to install all repository skills globally for the selected agent.
Installs: All skills installed directly from the dbt-labs/dbt-agent-skills GitHub repository through Tessl.
Before you start
Tessl
Most skills assume dbt is installed and configured, a dbt project containing dbt_project.yml, and basic familiarity with dbt concepts; fetching-dbt-docs and configuring-dbt-mcp-server can be used without an existing project.
For building the dbt Semantic Layer, the legacy specification requires dbt Core 1.6–1.11; the latest specification requires dbt Core 1.12+ or Fusion.
For local dbt MCP server setup, uv must be installed, and a dbt project is required for CLI commands. Remote setup instead requires a dbt Cloud account with AI features enabled, a production environment ID, and a personal access token or service token.
For cross-platform migration, dbt Fusion must be installed and working, and it must connect to both source and target platforms.
dbt v2 is required when using the included migrating-dbt-core-to-v2 skill.
Run this terminal command to install all skills directly from the GitHub repository with Tessl.
A specific documentation question replacing [QUESTION]
Access to fetch docs.getdbt.com pages
Check the repository for this skill’s setup.
Use the fetching-dbt-docs skill to find official dbt documentation about [TOPIC]. Search the documentation index first, return the relevant page URLs, then fetch the best matching page using its .md URL and briefly summarize the section that answers [QUESTION]. Treat [TOPIC] and [QUESTION] as placeholders for my subject and specific question.
Path to the dbt project containing dbt_project.yml
Model name replacing [MODEL]
The dbt CLI flavor, if known
A configured dbt connection
Check the repository for this skill’s setup.
Use the running-dbt-commands skill in my dbt project to preview up to 10 rows from [MODEL]. Ask which dbt CLI flavor I use if it is unclear, use an explicit selector, and do not run the entire project. Treat [MODEL] as the model-name placeholder.
Existing documented models from which to infer house style
Check the repository for this skill’s setup.
Use the maintaining-dbt-documentation skill to audit documentation coverage in [MODELS_FOLDER]. Read several documented models to infer the project's house style, then draft only the missing model and column descriptions in that one folder for my review. Do not commit or push. Treat [MODELS_FOLDER] as a project-relative folder placeholder.
Claude
Cursor
Copilot
Cline