← Back to Claude Skills

/createsolution

AI Board Review workflow for multi-perspective solution design

v2.0.0 - Critique Loop

Invoke this skill in Claude Code:

/createsolution Build a stock portfolio tracker

Add --skip-critique to bypass the critique loop

Workflow Position

1
/createsolution
You are here
2
/createplan
3
/executeplan
4
/gitlab-board

Multi-AI Perspectives

Three AI models (Claude, Gemini, Codex) independently analyze your requirements, then their best ideas are synthesized into a single consolidated solution. The optional critique loop challenges the result for over-engineering.

Full Workflow (11 Steps)

# Default: Full workflow with critique loop DESIGN PHASE (Steps 1-6) Step 1: Name approval (human confirms project name) Step 2: Refine & format the program ask Step 3: Claude's initial analysis Step 4: AI Board Review (Gemini + Codex, in parallel) Step 5: Synthesize v1 (merge best of 3 perspectives) Step 6: Optimize v1 # If --skip-critique: Jump to Step 11 CRITIQUE LOOP (Steps 7-10) Step 7: Critique Review (3 AI models challenge the solution) Step 8: Refactoring plan from critique consensus Step 9: Re-synthesize v2 incorporating feedback Step 10: Final optimize PRESENTATION Step 11: Executive HTML presentation on nginx

AI Board Review

Claude (Initial)

Analyzes requirements and proposes a primary solution with architecture, tech stack, and implementation approach.

Gemini (Design)

Focuses on high-level reasoning, system integration, user flows, and architectural coherence.

Codex (Technical)

Focuses on technical implementation feasibility, code structure, stack choices, and practical concerns.

Synthesis

Best components from all three are merged, conflicts resolved, producing a consolidated solution document.

Critique Loop (Optional)

Challenge Assumptions

After the initial solution, all three AI models switch roles from designer to critic. They evaluate for over-engineering, identify simpler alternatives, and flag red flags. The consensus drives a v2 that's leaner and more robust.

Gemini Critique

Rates complexity (1-10), proposes "boring technology" alternatives, identifies what to remove to cut 30%.

Codex Critique

Estimates lines of code, audits each dependency, scores deployment complexity, proposes concrete simpler tools.

Claude Self-Critique

Challenges own assumptions, identifies invented requirements, rates confidence, finds the "90% solution that's 50% simpler".

Artifacts Produced

Without Critique (--skip-critique): 6 files

With Critique (default): 11 files

Usage

# Full workflow with critique (11 steps)
/createsolution Build a stock portfolio tracker

# Skip critique loop (7 steps, faster)
/createsolution --skip-critique Build a stock portfolio tracker

# Output locations:
# Markdown: ~/projects/ainotes/shared/createsolution/
# HTML: nginx.ai-servicers.com/{username}/createsolution/{descriptor}/

Related Skills

/createplan

Takes the solution and creates a phased implementation plan with agent assignments.

/executeplan

Creates GitLab board from plan artifacts with trackable cards.

/gitlab-board

Board automation - processes ready cards, executes work, manages the kanban flow.