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

sdlc-agent-03-business-understanding

Generates a Business Understanding Document from validated requirements in a Databricks SDLC pipeline.

1|2|Updated Aug 10, 2026
Search Tags:#multi-agent#sdlc#databricks#genie#unity catalog#foundation model api#business understanding
One-click install
npx skills add https://github.com/TRRaveendra/AI-Agents-Databricks --skill sdlc-agent-03-business-understanding-trraveendra
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sdlc-agent-03-business-understanding
Source: https://github.com/TRRaveendra/AI-Agents-Databricks/tree/main/ai-agent-skills/databricks-sdlc-agent-skills/03-business-understanding-agent
Command: npx skills add https://github.com/TRRaveendra/AI-Agents-Databricks --skill sdlc-agent-03-business-understanding-trraveendra

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? In a multi-agent Databricks SDLC pipeline, validated requirements must be translated into a structured Business Understanding Document before downstream agents can discover metadata, build mappings, or write code. Doing this manually is slow and inconsistent, and ad-hoc LLM calls produce outputs that break the orchestrator's contract validation. ## Core Features & Use Cases - Business Understanding Generation: Articulates the business objective, current-state vs future-state process, business rules, KPIs, risks, assumptions, and deliverables from a validated requirement. - Contract-Compliant Output: Returns JSON matching a strict output contract (status, artifact_path, summary, citations, tool_calls, tokens) that the LangGraph supervisor validates before handoff. - Databricks Tool Integration: Calls the Foundation Model API for synthesis and the Genie Agents Conversation API for natural-language business-metric questions, with citations recorded. - Use Case: A Jira ticket flows through Agents 1-2 and arrives as a validated requirement; this agent reads it from the Unity Catalog volume, produces business_understanding.md/json, writes it to /Volumes/{catalog}/state/artifacts/{ticket}/, and logs the run to state.agent_runs. ## Quick Start Hand this agent a validated requirement from the upstream pipeline and ask it to produce the Business Understanding Document for the ticket.

Frequently Asked Questions about sdlc-agent-03-business-understanding

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

FAQPage Schema
How do I generate a business understanding document from a validated requirement in Databricks?▼

Read the validated requirement artifact from the Unity Catalog volume for the ticket, call the Foundation Model API to synthesize objectives, processes, rules, and KPIs, then write business_understanding.md/json back to the ticket's artifact path and return the contract JSON.

Can Genie Code be called headlessly from a Databricks agent or job?▼

No. Genie Code is UI-only with no public REST or SDK endpoint, so orchestrators cannot invoke it. Use the Foundation Model API on Model Serving or a Mosaic AI Agent endpoint for code and document generation instead.

What is the Genie Agents Conversation API used for in this pipeline?▼

It answers natural-language data questions against a Genie Space, such as current business metrics or reporting context. Agents use start_conversation_and_wait or create_message_and_wait and record the conversation_id and generated SQL as citations.

What happens when the agent output fails contract validation?▼

The agent gets one repair retry with the validation error appended to its context. If it still fails, the run is routed to state.needs_human so the approval-gate agents can review it.

When should the agent return needs_clarification instead of complete?▼

Return needs_clarification when required upstream context is missing or contradictory, including specific questions. Never invent metadata, business rules, or test results to fill gaps.

Related Skills

Explore More Software Engineering Skills→
ddieppaddieppa
1

jira-ticket-reader

Read Jira ticket data from the current git branch and generate Markdown task documentation.

Community
Advanced
DKSangDKSang
2

des-business-questions

Convert ambiguous business intent into a prioritized question catalog mapped to consumers and decisions.

Community
Advanced
victorlouvictorlou
4

create-tools

Create Databricks Genie spaces, Vector Search indexes, UC functions, and HTTP connections.

Community
Intermediate
yiwei79yiwei79

dynamic-full-auto

Automate multi-wave workflows with discovery, evidence insertion, and gated handoffs.

Official
Advanced