From abstract-to-concrete-design
Validate problem understanding, stakeholder alignment, context freshness, and gap analysis before proceeding to ideation. Run after design:feed (pre-agent) and after Wave 1 (post-research).
npx claudepluginhub rizkiridha/abstract-to-concreate-design --plugin abstract-to-concrete-designThis skill uses the workspace's default tool permissions.
You are validating that the design problem is well-understood, stakeholders are aligned, context is fresh, and gaps are addressed before proceeding.
Verifies tests pass on completed feature branch, presents options to merge locally, create GitHub PR, keep as-is or discard; executes choice and cleans up worktree.
Guides root cause investigation for bugs, test failures, unexpected behavior, performance issues, and build failures before proposing fixes.
Writes implementation plans from specs for multi-step tasks, mapping files and breaking into TDD bite-sized steps before coding.
Share bugs, ideas, or general feedback.
You are validating that the design problem is well-understood, stakeholders are aligned, context is fresh, and gaps are addressed before proceeding.
Pre-agent: After /design:feed, before /design:run
Post-research: After Wave 1 agents complete, before Wave 2
.design/DESIGN-STATE.md to determine current stateIf neither state matches:
"Nothing to validate right now. Run
/design:feedto add context, then validate. Or run/design:runto complete Wave 1, then validate."
Read .design/BRIEF.md to understand all context sections added.
Build stakeholder map from "Source:" labels:
MoM - Product Sync 2026-03-10 → Product Manager
Email from Engineering → Engineer
User interview with Sarah → Customer
Slack thread - Design Critique → Designer
Update BRIEF.md by appending:
## Stakeholders
| Source | Role | Key Input |
|--------|------|-----------|
| [source] | [role] | [1-line summary] |
Scan context sections for contradictory statements:
If conflicts found, add to validation report:
### Potential Conflicts
- [conflict description]
- Suggestion: Discuss with team before proceeding
Parse timestamps from context sections:
30 days: ⚠️ Potentially stale
Add to validation report:
### Context Age
- [date] ([age]) - [status]
Rate confidence by area:
| Area | Confidence | Notes |
|---|---|---|
| Problem statement | HIGH/MEDIUM/LOW | Based on source agreement |
| User needs | HIGH/MEDIUM/LOW | Based on research depth |
| Technical constraints | HIGH/MEDIUM/LOW | Based on engineering input |
| Business goals | HIGH/MEDIUM/LOW | Based on strategy docs |
Prompt designer:
## Problem Reflection
Before we run agents, confirm your understanding:
**What problem are we solving?**
[Synthesize from BRIEF.md]
**Who experiences this?**
[Users identified in context]
**If we solve this, what changes?**
[Designer answers]
**If we DON'T solve this, what happens?**
[Designer answers]
---
Ready to proceed? (yes/no/revise)
Save designer's responses to .design/PROBLEM-REFLECTION.md.
Initialize sign-offs in DESIGN-STATE.md:
## Sign-offs
- [ ] Product Manager - pending
- [ ] Engineering - pending
- [ ] Designer - pending
Read .design/research/RESEARCH.md to get the synthesized problem statement.
Read .design/PROBLEM-REFLECTION.md (from pre-agent validation).
Compare:
## Research Validation
Research has synthesized:
**[Problem Statement from RESEARCH.md]**
**Your earlier reflection:**
[From PROBLEM-REFLECTION.md]
**Quick Check:**
- [ ] Research matches my understanding
- [ ] This problem is worth solving (impact > effort)
- [ ] We have enough evidence to proceed
---
If all pass → proceed to Wave 2
If any fail → /design:feed more context
Update confidence based on research findings:
## Understanding Confidence (Post-Research)
| Area | Confidence | Evidence |
|------|------------|----------|
| Problem statement | HIGH | Research synthesis confirms |
| User needs | MEDIUM | Research adds detail |
| [etc] | [rating] | [source] |
From research, surface unknowns that could change the solution:
### Critical Unknowns
- What happens if export fails? (edge case)
- How often is this used? (prioritization)
- Technical capacity? (scope)
Create .design/VALIDATION-REPORT.md:
# Validation Report: [PROJECT]
Generated: [DATE]
## Status
- [x] Pre-agent validation: [date]
- [ ] Post-research validation: pending
## Stakeholders
[Stakeholder table]
## Conflicts
[Any conflicts detected]
## Context Health
[Age analysis]
## Confidence Score
[Confidence table]
## Critical Gaps
[Unknowns]
## Sign-offs
[Who approved]
## Next Step
[Proceed / Need more context / Revise]
Update .design/DESIGN-STATE.md:
.design/BRIEF.md (stakeholder section).design/GAPS.md (confidence scoring).design/DESIGN-STATE.md (validation status).design/PROBLEM-REFLECTION.md (designer's answers).design/VALIDATION-REPORT.md (full report)/design:validate
Claude Code, OpenCode CLI