Engineering
BDD-prompts-with-eval avatar

BDD-prompts-with-eval

An AI-driven active listening framework to extract, clarify, and structure requirements, business values, and scope from ambiguous user stories.

Introduction

This skill acts as a structured conversational facilitator, designed for software engineers, product managers, and business analysts who need to refine ambiguous ideas into actionable development specifications. By utilizing techniques rooted in Carl Rogers’ Active Listening, the Toyota 5 Whys, and Socratic questioning, it prevents the common pitfall of jumping prematurely into technical solutions. Instead, it forces a rigorous discovery phase to uncover hidden pain points, identify clear user personas, and establish concrete scope boundaries (In-Scope, Out-of-Scope, and TBD items).

The process follows a systematic workflow: initial understanding, deep exploration using 5W1H and Socratic questioning, persona definition, scope boundary confirmation, and final hypothesis verification. This ensures that every software requirement is backed by a verifiable business goal and clear success metrics. It is particularly effective for teams using BDD (Behavior Driven Development) or Example Mapping, serving as the essential precursor to technical design.

  • Employs open-ended questions to avoid leading stakeholders and encourage honest, detailed requirements discovery.

  • Uses the 5 Whys technique to drill down into the root cause of a requested feature, ensuring only high-value functions reach the implementation stage.

  • Facilitates explicit scope boundary creation, clearly identifying what is excluded to prevent project creep.

  • Generates structured, markdown-formatted outputs including business values, KPI indicators, primary/secondary personas, and blocker identification.

  • Guides the user through a non-linear but disciplined dialogue process, prioritizing active verification and paraphrasing to eliminate misunderstandings.

  • Best suited for complex, brownfield, or highly ambiguous software projects where user intent is often obscured.

  • Integrates seamlessly with existing BDD and Agile workflows, specifically acting as the upstream bridge to Example Mapping.

  • Input expected: Vague user stories, raw stakeholder interviews, incomplete feature ideas, or rough notes.

  • Output provided: A comprehensive requirement summary, a prioritized list of user stories, and a roadmap for technical validation.

  • Constraints: Recommended session duration is 20-30 minutes; for overly complex requirements, it suggests breaking stories down further.

Repository Stats

Stars
0
Forks
0
Open Issues
0
Language
TypeScript
Default Branch
main
Sync Status
Idle
Last Synced
May 3, 2026, 07:54 PM
View on GitHub