skills.rest
TRRaveendra
TRRaveendraCOMMUNITY
Explore all skills by this author →
GitHub

sdlc-agent-06-data-model

Generates Bronze/Silver/Gold data models and Delta DDL scripts from mapping documents on Databricks.

1|2|Updated Aug 10, 2026
Search Tags:#multi-agent#data modeling#medallion architecture#databricks#ddl#unity catalog#delta lake
One-click install
npx skills add https://github.com/TRRaveendra/AI-Agents-Databricks --skill sdlc-agent-06-data-model-trraveendra
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sdlc-agent-06-data-model
Source: https://github.com/TRRaveendra/AI-Agents-Databricks/tree/main/ai-agent-skills/databricks-sdlc-agent-skills/06-data-model-agent
Command: npx skills add https://github.com/TRRaveendra/AI-Agents-Databricks --skill sdlc-agent-06-data-model-trraveendra

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires databricks-sdk, and includes references (resource) components.

What problem does it solve? Designing medallion-architecture data models and writing validated Delta DDL by hand is slow and error-prone in a multi-agent SDLC pipeline. This Skill acts as Agent 6 of 15, converting an upstream Mapping Document into a governed Data Model plus DDL scripts that downstream agents can consume. ## Core Features & Use Cases - Medallion Data Modeling: Designs Bronze/Silver/Gold layer models with dimension/fact tables, constraints, and keys. - DDL Generation & Validation: Produces Delta table DDL with partitioning, liquid clustering, and Unity Catalog naming, then dry-run validates it on a SQL warehouse before finalizing. - Contract-Driven Handoff: Returns JSON matching a strict output contract, writes artifacts to /Volumes/{catalog}/state/artifacts/{ticket}/, and logs runs to state.agent_runs for audit. - Use Case: In a ticket-driven pipeline, the Mapping Document Agent finishes, and this agent automatically drafts the warehouse schema, requests human approval, and hands validated DDL to the PySpark Development Agent. ## Quick Start Given the mapping document for ticket PROJ-123 in the artifacts volume, act as the Data Model Agent and produce the data model and DDL scripts.

Frequently Asked Questions about sdlc-agent-06-data-model

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate Delta table DDL from a mapping document on Databricks?▼

Feed the mapping document to this agent, which calls the Databricks Foundation Model API to draft DDL with partitioning, liquid clustering, and Unity Catalog naming. It then validates the DDL with a dry-run CREATE on a scratch schema before writing ddl/*.sql and data_model.json to the artifacts volume.

Can I call Genie Code from a Databricks job or orchestrator via API?▼

No. Genie Code is UI-only with no public headless API. Use the Foundation Model API (Model Serving pay-per-token endpoints) or a Mosaic AI Agent endpoint for code generation, and reserve the Genie Agents Conversation API for natural-language data questions.

What output format does the Data Model Agent return to the orchestrator?▼

It returns JSON with agent id, ticket_id, status (complete, needs_clarification, or failed), artifact_path, summary, citations, tool_calls, token counts, and duration. The LangGraph supervisor validates this contract and allows one repair retry on mismatch.

Does this agent require human approval before downstream agents run?▼

Yes. The data model stage is an approval gate: after writing artifacts and upserting state.artifacts and state.agent_runs rows, unresolved validation failures route to state.needs_human so reviewers can approve before the PySpark Development Agent proceeds.

What happens when upstream artifacts are missing or contradictory?▼

The agent returns status needs_clarification with specific questions instead of inventing metadata or business rules. This keeps the pipeline grounded in retrievable citations such as UC tables, prior artifact paths, or Genie conversation ids.

Related Skills

Explore More Software Engineering Skills→
snoodleboot-iosnoodleboot-io

data-model-discovery

Gathers and validates data model requirements through structured discovery questions before schema design.

Official
Intermediate
ancolemanancoleman
503

transforming-data

Transform raw data into analytical assets using ETL/ELT patterns and orchestration tools.

Community
Advanced
tja688tja688

unity-catalog-tagger

Apply and manage Unity Catalog metadata tags for tables, columns, and schemas.

Community
Advanced
beelabstudiobeelabstudio

dbt-specialist

Develops and optimizes dbt data transformation pipelines using dimensional modeling and analytics engineering practices.

Official
Basic