Back to browse
GitHub Repository

Audit a dbt project for what an AI agent will get wrong if you point it at the data today.

4 starsPython

A skill to audit your dbt project for what an AI agent will get wrong

by matthieu_bl·Jun 11, 2026·3 points·0 comments

AI Analysis

●●SolidNiche GemSolve My Problem

Catches AI-breaking dbt issues like conflicting revenue metrics and YAML/SQL mismatches.

Strengths
  • Specific to AI agent failures, not general dbt linting or testing
  • Ships SQL verification queries for each finding so you can validate manually
  • Uses sqlglot for actual SQL parsing instead of regex pattern matching
Weaknesses
  • Requires Claude Code Skills support, limiting which AI assistants can run it
  • Needs dbt compile to run first for accurate macro resolution
Category
Target Audience

Data engineers and analytics teams using dbt with AI coding assistants

Similar To

dbt-core tests · dbt-expectations · elementary-data

Similar Projects

AI/ML●●●Banger

A Claude Code skill that tells Claude when not to use Claude

Traffic-light audit system beats vague 'optimize your LLM spend' advice from competitors.

Big BrainSolve My Problem
metravod
1028d ago