What problem does it solve?
Manually writing MetricFlow semantic model YAML is time-consuming, error-prone, and often fails validation, leading to broken metrics and inconsistent data definitions across data teams.
Core Features & Use Cases
- Automated Semantic Modeling: Generates production-ready MetricFlow YAML directly from database table schemas, eliminating manual YAML writing effort.
- Iterative Validation & Fixing: Runs built-in validation checks and automatically fixes YAML errors until the model passes all requirements.
- Knowledge Base Publishing: Automatically publishes validated semantic models to the central Knowledge Base for reuse across all data tools and agents.
- Use Case: A data engineer can turn a set of raw sales and user database tables into a fully validated, reusable semantic model in minutes instead of hours of manual work, ensuring all downstream metrics are consistent and accurate.
Quick Start
Use the metricflow-semantic-authoring skill to generate a validated semantic model for the public.sales fact table and publish it to the Knowledge Base.