What problem does it solve?
This Skill eliminates the risk of manual, error-prone full rewrites of Bronze ingestion artifacts when Low-Level Design (LLD) or Story Technical Model (STM) specifications change, ensuring data pipelines stay aligned with design updates without disrupting active runs.
Core Features & Use Cases
- 8 Targeted Update Scenarios: Handles adding/removing/renaming Bronze tables, adjusting per-table ingestion knobs (retries, timeouts, empty-input behavior), updating runner/factory logic, refreshing data quality rules, updating Spark expectations runners, and generating plain SQL Bronze DDL migrations.
- Minimal Incremental Edits: Only modifies the specific files impacted by design changes, preserving all unrelated ingestion code and configuration to avoid breaking existing pipeline functionality.
- Built-in Validation: Automatically enforces project-specific coding patterns, library version requirements, and test coverage, with mandatory linting and test runs before completion.
- Real-World Use Case: For example, if the LLD adds a new patient_vitals Bronze table, the skill automatically creates the corresponding YAML config, updates the ingestion factory if needed, adds required test coverage, and generates the matching Delta Lake DDL migration without touching unrelated ingestion components.
Quick Start
Use the update-ingestion skill to add the new patient_vitals Bronze table defined in the latest LLD.