skills.rest
rizqinrr
rizqinrrCOMMUNITY
Explore all skills by this author →
GitHub

get-prd

Interviews users through structured rounds and produces an approved PRD document.

665|3|Updated Sep 12, 2026
Search Tags:#planning#specification#requirements#product management#discovery#interview#prd
One-click install
npx skills add https://github.com/rizqinrr/viserys-agent --skill get-prd-rizqinrr
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: get-prd
Source: https://github.com/rizqinrr/viserys-agent/tree/main/skills/get-prd
Command: npx skills add https://github.com/rizqinrr/viserys-agent --skill get-prd-rizqinrr

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Vague feature requests lead agents to build the wrong thing based on silent assumptions. This Skill turns an underspecified ask into an approved PRD through a structured interview, so downstream planning starts from a verified shared contract instead of guesses. ## Core Features & Use Cases - Design-tree interviewing: Maps every decision as a tree, asks the current frontier in numbered rounds, and attaches a recommendation to each question so the interview converges instead of rambling. - Intent extraction with confidence tracking: Records a hypothesis with an explicit confidence percentage, probes want-vs-should-want answers, and restates intent in the user's own words before writing anything. - PRD authoring with an approval gate: Produces a complete PRD saved to docs/prd/<name>.md covering problem statement, user stories, technical decisions, boundaries, and success criteria, gated on an explicit user yes. - Use Case: A user says "build me a dashboard for our metrics." The skill runs hypothesis seeding and round-based questions, discovers the real need is a personal experiment tracker, and writes an approved PRD ready for task breakdown. ## Quick Start Ask the agent to run get-prd and interview you toward a PRD for the new feature or system you want to build.

Frequently Asked Questions about get-prd

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

FAQPage Schema
How do I turn a vague feature idea into a PRD with an AI agent?▼

Run a structured interview workflow that maps decisions as a tree, asks the current frontier in numbered rounds with recommendations, then writes the PRD. The get-prd skill does this in one pass and saves the result to docs/prd/<name>.md after your explicit approval.

What sections should a product requirements document include?▼

This PRD template requires problem statement, solution, user stories, objective, technical decisions, project structure, commands, testing strategy, boundaries, out of scope, success criteria, and open questions. Every section must be filled or explicitly marked not applicable with a reason.

When should I not use an interview-based PRD workflow?▼

Skip it for bug fixes, renames, mechanical changes, pure information requests, and small amendments to an existing spec. It also requires a live responsive user, so non-interactive contexts like CI pipelines should flag underspecified asks as blockers instead.

How do I handle a feature request that bundles multiple capabilities?▼

Propose a capability map first: a small table of module ids, responsibilities, dependencies, and build order, gated on user approval. Then run the interview and PRD process once per module in build order rather than writing one monolithic PRD.

Why does the interview attach a recommendation to every question?▼

Users react to a wrong recommendation faster than they generate answers from scratch, and it surfaces the agent's assumptions for correction. The recorded acceptances and overrides later feed the PRD's Technical Decisions section.

Related Skills

Explore More Product & Management Skills→
Martino17xMartino17x

brainstorming

Transform rough ideas into structured designs through guided questioning and alternative exploration.

Community
Basic
andyzengmathandyzengmath
24

ql-spec

Generate machine-verifiable PRDs with user stories and acceptance criteria.

Community
Advanced
Handy369Handy369

api-and-interface-design

Define contract shapes and boundary semantics for API and module interfaces before coding.

Community
Intermediate
whw23whw23

speckit-companion-mark-complete

Marks the active spec-kit feature as completed in .spec-context.json via a Python script.

Community
Basic