erpaval
Workflow for autonomous software development with Claude Code. Classifiers route scope, complexity, directory state, and spec…
Optional path to raw content (brief, notes, tickets) or a one-line topic.
Claude Code
$ npx -y skills add laithalsaadoon/erpaval --skill product-discovery --agent claude-code
Codex
$ npx skills add laithalsaadoon/erpaval --skill product-discovery -a codex -y
Cursor
$ npx skills add laithalsaadoon/erpaval --skill product-discovery -a cursor -y
Gemini CLI
$ npx skills add laithalsaadoon/erpaval --skill product-discovery -a gemini-cli -y
GitHub Copilot
$ npx skills add laithalsaadoon/erpaval --skill product-discovery -a github-copilot -y
OpenCode
$ npx skills add laithalsaadoon/erpaval --skill product-discovery -a opencode -y
Windsurf
$ npx skills add laithalsaadoon/erpaval --skill product-discovery -a windsurf -y
name: product-discovery
description: >
File-first orchestrator for product discovery and spec frameworks.
Frames the problem with a Discovery Lead / Product Analyst / System
Architect role, spawns parallel framework agents (Double Diamond,
How Might We, Jobs to Be Done, user stories + INVEST, EARS for
AI-coder specs, Gherkin), synthesizes into a PRD or discovery memo,
runs a critic pass. Absorbs prd-drafter (idea to PRD via 3 parallel
research roles) and design-thinking (pain-point excavation, landscape
research, brainstorm). Ships per-framework references, role prompts,
templates for PRDs, HMW outputs, job stories, EARS specs. Use when
the user asks to write a PRD, draft requirements, run discovery,
brainstorm a product, turn insights into HMW questions, write job
stories, spec a feature in EARS for Kiro or Spec Kit, define user
stories with INVEST, run a Double Diamond round, or mentions JTBD,
acceptance criteria, Kiro spec format. Do NOT use for business
strategy (Rumelt, Wardley, PR-FAQ as leadership artifact) — use
product-strategy.
arguments:
- name: source
description: Optional path to raw content (brief, notes, tickets) or a one-line topic.
required: false
user_facing: true| Reference | When to load | | -------------------------------------------- | -------------------------------------------------------------------------------------------------------- | | `references/orchestrator.md` | Running the pipeline — 5 routes, parallel Phase 2, critic review, deliver | | `references/write-protocol.md` | Canonical write-protocol block copied into every subagent prompt | | `references/frameworks/INDEX.md` | Routing guide — discovery + spec decision tables, discovery → spec composition | | `references/frameworks/*.md` | Per-framework files — double-diamond, how-might-we, jtbd-job-stories, user-stories-invest, ears, gherkin | | `references/methodology/discovery-rounds.md` | Six-phase discovery methodology | | `references/inference-heuristics.md` | Signal-to-inference tables for PRD-route Intent Profile capture | | `references/quality/prd.md` | Per-section PRD quality bar, cross-section consistency checks | | `references/roles/product-analyst.md` | PRD route — vision, personas, user stories, IA, milestones (sections 1-6, 12) | | `references/roles/system-architect.md` | PRD route — NFRs, data models, API contracts, edge cases (sections 7-10) | | `references/roles/research-scout.md` | PRD route — competitive landscape, open questions (sections 11, 13) | | `references/roles/discovery-lead.md` | Discovery-round route — pain-point excavation, Double Diamond cadence | | `references/roles/hmw-framer.md` | HMW-only route (also hard-dep entry point for erpaval CL-RIGOR fuzzy) | | `references/roles/jtbd-interviewer.md` | JTBD-only route — job stories from PM-provided source material | | `references/roles/ears-specifier.md` | EARS-only route (also hard-dep entry point for erpaval CL-RIGOR contract-unclear) | | `references/roles/prd-synthesizer.md` | PRD route Phase 3 — composes final PRD from three parallel work logs | | `references/roles/discovery-critic.md` | Phase 3.5 critic — rubric-graded review of any synthesized artifact | | `templates/prd-template.md` | 15-section PRD skeleton with `[FILL]` markers | | `templates/worklog-skeleton.md` | Per-role work-log skeleton with embedded write-protocol | | `templates/discovery-memo.md` | Discovery-round output shape | | `templates/double-diamond-workbook.md` | Double Diamond four-stage workbook | | `templates/hmw-skeleton.md` | HMW brainstorms file — `brainstorms/NNN-<slug>-requirements.md` | | `templates/jtbd-skeleton.md` | JTBD job-stories file | | `templates/ears-spec-skeleton.md` | EARS spec file — `specs/NNN-<slug>/spec.md` |
One skill, five routes. Takes a problem at any maturity (one-sentence idea, research notes, fuzzy framing, clear feature) and routes to the right framework pipeline. File-first write protocol throughout. Every role is a Markdown prompt copied verbatim into a `general-purpose` `Agent` call — zero custom agent files.
flowchart TD A[User brief] --> B[Phase 0: Route] B -->|PRD ask| P1[Phase 1: Intent Profile] B -->|Discovery round| D1[Phase 1: discovery-lead framing] B -->|HMW only| H1[Phase 1: framing] B -->|EARS only| E1[Phase 1: framing] B -->|JTBD only| J1[Phase 1: framing] P1 --> P2[Phase 2: product-analyst + system-architect + research-scout in parallel] D1 --> D2[Phase 2: hmw-framer / jtbd-interviewer / research-scout as routed] H1 --> H2[Phase
Autonomous software development for Claude Code. ERPAVal stands for Explore · Research · Plan · Act · Validate — the five-phase loop, plus a sixth Compound phase that writes durable lessons to disk so the next session inherits what this one learned.
Workflow for autonomous software development with Claude Code. Classifiers route scope, complexity, directory state, and spec…
Shared canonical references for product-design frameworks — Pyramid Principle, Working Backwards, Double Diamond, research…
Agent UX patterns and human-in-the-loop design: Twilio A2H protocol, Levels of Autonomy (L0-L5), inbox pattern, progressive…
Customer research and problem framing using the research-design discipline: hypothesis, null hypothesis, MECE questions, methods,…
Audit and rewrite prompts against Claude Opus 4.7 instruction-following best practices. Produces a scored audit plus a rewritten…
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…
Ask which skill or flow fits your situation. A router over the skills in this repo.