design

name	design
description	Convert specification into technical design. Use when creating architecture, defining components, interfaces, and data flows.
license	MIT
compatibility	Claude Code, GitHub Copilot, OpenCode, Cursor, Codex, Gemini CLI
metadata	{"author":"paruff","suite":"uFawkesAI"}

Skill: Design

Load trigger: "load design skill" > DORA: Cap 3 (AI-Accessible Internal Data) Token cost: Low

Purpose

Convert specification into technical design.

Responsibilities

• Decompose architecture
• Identify components
• Define interfaces
• Model data flows
• Validate K8s design
• Document tradeoffs

Sub-Skills

Skill	Purpose
design/architecture-decomposition	Break down system architecture
design/component-identification	Identify required components
design/interface-definition	Define component interfaces
design/k8s-design-validation	Validate Kubernetes design

Dependencies

Skill	Relationship
spec	Consumes specification and requirements

Inputs

• specification.md (from spec)
• Existing architecture (if updating)
• Policy documents

Outputs

• design.md
• architecture-decisions.md
• component-interfaces.md

Design Rules

Architecture

• Follows established patterns
• Aligns with existing architecture
• Scalability considered
• Security embedded (not bolted on)

Components

• Single responsibility
• Clear boundaries
• Loose coupling
• High cohesion

Interfaces

• Contracts defined
• Versioning considered
• Error handling specified
• Input validation defined

Tradeoffs

• Alternatives documented
• Rationale provided
• Risks identified
• Mitigation strategies defined

Output Format

{
  "skill": "design",
  "status": "pass | fail",
  "architecture": {
    "overview": "description",
    "components": ["auth-service", "user-api", "database"],
    "patterns": ["microservices", "event-driven"]
  },
  "interfaces": [
    {
      "component": "auth-service",
      "inputs": ["email", "password"],
      "outputs": ["token", "user-id"]
    }
  ],
  "tradeoffs": [
    {
      "decision": "Use PostgreSQL",
      "alternatives": ["MySQL", "MongoDB"],
      "rationale": "ACID compliance required"
    }
  ]
}

Success Criteria

• Architecture is clear and documented
• Components are well-defined
• Interfaces are explicit
• Tradeoffs are justified