prd-v01-problem-framing_skill

This skill converts vague product ideas into evidence-anchored problem statements with CFD IDs and a structured problem table for PRD v0.1.
  • Python

17

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill mattgierhart/prd-driven-context-engineering --skill prd-v01-problem-framing

  • SKILL.md4.4 KB

Overview

This skill transforms vague product ideas into evidence-anchored problem statements for early PRD v0.1 work. It converts market signals and user quotes into structured problem tables with CFD evidence IDs and quantified costs. The output is testable, prioritized, and ready for v0.2 market definition.

How this skill works

It runs a five-step workflow: assess gaps, anchor evidence with CFD entries, extract multiple pain dimensions from each source, quantify time/money/risk costs, and draft the problem statement table. Each claim must reference one or more CFD-IDs and meet quality gates (Tier 1–2 evidence, quantified cost, specific who, why-now hypothesis). The skill refuses to draft if too many elements are missing and recommends targeted research instead.

When to use it

  • Starting a new product or feature (PRD v0.1)
  • Validating market opportunity or drafting the PRD Why section
  • When asked to "frame the problem", "define pain points", or "write problem statement"
  • After collecting user quotes, invoices, or workaround evidence
  • Before committing engineering or design effort to solutions

Best practices

  • Require at least two validated elements before drafting; if three or more are missing, run deep research first
  • Create a CFD entry for every distinct pain and rank evidence by Tier (buying behavior highest)
  • Extract multiple problem dimensions from each quote—one quote can yield several problems
  • Quantify costs numerically (hours/week, dollars spent, revenue missed, probability-weighted risk)
  • Keep the Who specific and findable so you can source 10 prospects in 48 hours
  • Use quality gates: Tier 1–2 evidence present, cost quantified, and a testable why-now hypothesis

Example use cases

  • Convert customer support transcripts and invoices into a table of 3–5 prioritized problems with CFD-IDs
  • Turn a vague feature ask from sales into a validated problem statement ready for PRD v0.1
  • Assess market fit for a proposed feature by mapping evidence tiers and quantifying opportunity cost
  • Draft the PRD Why section with explicit persona, pain, cost, urgency, and impossible outcomes
  • Triage product backlog items by replacing feature descriptions with evidence-backed problem rows

FAQ

A CFD-ID is a compact evidence record linking a quote or data point to a pain dimension; it creates traceability so every claim can be verified against source evidence.

When will the skill refuse to draft a problem statement?

If three or more core elements (Who, What, Cost, Why now, What's impossible) are missing or only speculative, the skill will recommend targeted research before drafting.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational