teach

To extract reusable coding patterns, provide a reference artifact path like @models/staging/stg_orders.sql. The system analyzes structure, naming conventions, boilerplate, and data flow, then saves the results as a pattern with proper source references for AI training.

Pattern extraction works across multiple file types including models, queries, and configuration artifacts. It identifies structural patterns like file organization, naming conventions, boilerplate, and common blocks to standardize pattern-based learning for your projects.

Pattern-based learning works by analyzing a concrete example file to identify your codebase's structural and naming conventions. It then saves these extracted patterns using a standardized format via training_save, enabling the AI to apply them to future missions.

You do not need to manually specify coding patterns. Simply provide the path to a reference artifact, and the system automatically identifies and extracts reusable structural patterns, boilerplate, and data flow logic from the file itself.

You can save extracted code patterns for future AI missions. The system uses a standardized format via training_save to store the identified structural patterns, ensuring they are properly referenced and available for subsequent automated tasks.

The best way to teach naming conventions and boilerplate is to provide a concrete example file. The system automatically extracts these structural patterns and saves them as a standardized pattern, allowing the AI to learn your codebase's specific structure.