Index/ Claude Skills/ PRD Skill — Stages & Information Flow
Claude skill · Socratic discovery

PRD Skill — Stages & Information Flow

Socratic discovery across four depth levels, seven phases, and six reference files to produce a build-ready requirements document. Grounded in the PuRDy methodology: platform-agnostic, quantification-forward, failure-pattern-aware.

01

The discovery flow

The reality check routes the request to one of four depth paths — each path runs the same phases at a deeper cut. Every phase is its own node, exactly as the skill draws it.

MINIMAL LIGHT STANDARD DEEP START — user asks for a PRD Step 0 · confirm scale · Step 0b · deployment target REALITY CHECK speed vs. complexity — pick a path Phase 0 · skip Phase 1 · 2–3 questions Phase 4 · 1 success metric Minimal PRD Phase 0 · silent classify Phase 1 · 3–5 min Phase 4 · 1–2 metrics Phase 7 · Light PRD Light PRD Phase 0 · classify & confirm Phase 1 · 7–10 min, 5 Whys Phase 2 · stakeholders Phase 3 · systems inventory Phase 4 · 2–3 metrics Phase 5 · top 3–4 risks Phase 6 · prioritization Phase 7 · [visual diagram?] Full PRD Phase 0 · full classify Phase 1 · 10–15 min, +3M Phase 2 · full map + RACI Phase 3 · full systems Phase 4 · + out-of-scope Phase 5 · all 7 patterns Phase 6 · prioritization Phase 7 · [visual diagram?] Full PRD+ visual diagram Offer build prompt & scaffold repo git init · docs/ · README → PRD complete

Each depth path runs the same phases at a deeper cut. Phase-by-phase detail follows.

02

Phase reference

PHASE 0Context &
classification
Applies: Light · Standard · Deep   initiative-taxonomy.md

5 initiative types: Data Integration, Process Automation, Tool Selection, Cross-Functional, Reporting/Analytics. Signal-phrase matching, hybrid detection, gap detection by type. Standard/Deep confirm classification with the user; Light notes it silently.

PHASE 1Problem
deep-dive
Applies: all depths   a3-framework.mdscale-effort.mdquestion-banks.md

Current-process mapping, pain points, scale quantification, 5 Whys root-cause analysis, 3M waste classification (Muda / Mura / Muri). Depth decides how deep: Minimal = 2–3 questions, Light = 3–5 min, Standard = 7–10 min with 5 Whys on the primary pain, Deep = full 5 Whys on every pain point + full 3M.

PHASE 2Stakeholder
map
Applies: Standard · Deep   question-banks.md

Role + decision-power scoring (1–5), resistance analysis, maintainer identification, missing-voices check. Standard covers key stakeholders and approvers. Deep adds full RACI, power/interest scoring, and an explicit resistance strategy for every potential blocker.

PHASE 3System & data
inventory
Applies: Standard · Deep (MVP: quick list only)   question-banks.md

Each system’s role, integration type (API / manual / file transfer), data sensitivity / PII flags, what can’t change, integration-complexity score. Deep adds data-ownership mapping, quality assessment, and transformation-complexity scoring.

PHASE 4Success
criteria
Applies: ALL depths — non-negotiable   scale-effort.md

Quantified metrics (current → target), MVP scope boundaries, explicit out-of-scope list. Standard/Deep add a pilot plan and timeline. This is the one phase that never gets skipped — even Minimal asks “what does success look like?”

PHASE 5Risks &
dependencies
Applies: Standard · Deep   failure-patterns.mdscale-effort.md

Risk + mitigation pairs, timeline / budget constraints, dependency map. 7-pattern failure scan: Integration Complexity, Ghost Town Build, Pre-Scale Paralysis, Cost Avalanche, Feature Gravity Well, Technical Debt Spiral, Reality Distortion. Standard scans top 3–4; Deep scans all 7. The 40-hour threshold routes large initiatives to formal approval.

PHASE 6Opportunity
prioritization

Impact / effort matrix on the ranked opportunities.

PHASE 7Generate
PRD

Build the full or light requirements document from the gathered inputs. Optionally produce a visual diagram first.

03

Depth levels at a glance

Four escalation levels controlled by project type — the skill scales depth to match scope.

Minimal — Quick Demo
Scale: Kiddie Pool, S effort
Phases: 0 skip, 1 (2–3q), 4 (1 metric)
Template: Minimal PRD
Time: 2–3 min

A throwaway sketch.

Light — MVP
Scale: Kiddie Pool, S–M effort
Phases: 0 silent, 1 abbrev., 4, 7
Template: Light PRD
Time: 5–8 questions

A buildable first version.

Standard — Internal Tool
Scale: Olympic Pool, M–L effort
Phases: all at moderate depth
Template: Full PRD
Time: 20–30 min

A real internal product.

Deep — Platform / Enterprise
Scale: Ocean, L–XL effort
Phases: all at full depth
Template: Full PRD + visual diagram
Time: 30–45 min

A platform bet.

04

Reference files

Six reference files pulled in by phase. The skill reads prd-template.md before starting and scans the others during the conversation.

Reference fileUsed byContent
prd-template.mdPhase 7Full and Light PRD templates with all sections.
initiative-taxonomy.mdPhase 05 initiative types, signal phrases, failure patterns per type, success benchmarks.
question-banks.mdPhases 1, 2, 3Context-specific questions: internal tools, stakeholder, training, workflow integration.
a3-framework.mdPhase 1Current → Target → Gap → Root Cause → Countermeasures structure.
failure-patterns.mdPhase 57 failure patterns with warning signs, prevention strategies, cascade models, immunity scorecard.
scale-effort.mdPhases 1, 4, 5Kiddie / Olympic / Ocean scale, S/M/L/XL effort tiers, 40-hour threshold, impact/effort matrix.
Anti-patterns the skill guards againstUsing Deep mode for a Kiddie Pool project — the most common failure · jumping to solutions in Phase 1 · accepting “a lot” / “better” / “faster” without numbers · a PRD with zero open questions · silently filling gaps in user requirements · assuming cloud for local-only projects · skipping repo scaffolding.