dev3-prd

Generate clarifying questions and write a pyramid-structured PRD from feature requests.

Updated Apr 12, 2026
One-click install
npx skills add https://github.com/Srajangpt1/dataeye --skill dev3-prd-srajangpt1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dev3-prd
Source: https://github.com/Srajangpt1/dataeye/tree/main/packages/browseros-agent/.claude/skills/dev3-prd
Command: npx skills add https://github.com/Srajangpt1/dataeye --skill dev3-prd-srajangpt1

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Many feature requests lack the clarity a staff engineer needs to produce an actionable design spec. This Skill generates the right clarifying questions, attempts to resolve them from the codebase and available artifacts, and then produces a concise, pyramid-structured Product Requirements Document (PRD) ready for implementation.

Core Features & Use Cases

  • Generate targeted questions based on the provided feature context, code exploration, and chosen design.
  • Self-resolve via repository artifacts by reading .llm/$ARGUMENTS/tmp_context.md, tmp_exploration.md, and design.md and marking answers as [RESOLVED] when possible.
  • Human-in-the-loop clarification by presenting only unresolved questions for user input and updating the question file with answers.
  • Produce a pyramid PRD organized into Executive Summary, Component Details, and Implementation Details, then write the final spec to .llm/$ARGUMENTS/prd.md and hand off to the implementation step.
  • Use Case: Turn a brief feature request and exploratory notes into a complete PRD for engineering handoff.

Quick Start

Use the dev3-prd skill to generate clarifying questions and draft a pyramid-style PRD for the feature "improve paste detection".

Frequently Asked Questions about dev3-prd

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

FAQPage Schema
How do I generate clarifying questions for a product requirements document?

To generate clarifying questions for a product requirements document, the Skill analyzes existing feature requests and design artifacts, then outputs unresolved unknowns marked as NEEDS INPUT for human review. It attempts to resolve questions automatically using repository context before soliciting user input.

What is the pyramid principle for structuring a PRD?

The pyramid principle for structuring a PRD organizes information hierarchically into an Executive Summary, Component Details, and Implementation Details. This approach synthesizes feature context and design artifacts into a concise, actionable spec ready for engineering handoff.

How do I turn feature exploration notes into a formal design doc?

You can turn feature exploration notes into a formal design doc by processing temporary context and exploration files alongside chosen designs. The Skill synthesizes these artifacts to produce a structured PRD and automatically triggers the implementation workflow upon completion.

Can I automatically resolve PRD unknowns from my codebase?

Yes, you can automatically resolve PRD unknowns from your codebase. The Skill reads temporary context, exploration findings, and design files to answer clarifying questions, marking them as RESOLVED when sufficient information is found in the repository artifacts.

Does this PRD generation workflow support human-in-the-loop clarification?

Yes, this PRD generation workflow supports human-in-the-loop clarification. After attempting to self-resolve unknowns from repository artifacts, it presents only the remaining unresolved questions to solicit necessary human answers before finalizing the product requirements document.

What are the limitations of automated PRD generation from feature requests?

A limitation of automated PRD generation is its dependency on pre-existing exploration files and design artifacts within the .llm path. If the codebase lacks sufficient context, the Skill cannot auto-resolve unknowns and will require manual human input for unanswered clarifying questions.