planning-phase

marcusgoll's avatarfrom marcusgoll

Standard Operating Procedure for /plan phase. Covers research depth, code reuse detection, design pattern selection, and architecture planning.

⭐0stars🔀0forks📁View on GitHub🕐Updated Nov 24, 2025
[Architecture Design]

When & Why to Use This Skill

This Claude skill provides a comprehensive Standard Operating Procedure (SOP) for the software development planning phase. It streamlines architecture design, research depth, and code reuse detection to ensure robust implementation plans. By integrating project documentation and defining API contracts early, it minimizes technical debt and improves development efficiency through structured research and complexity estimation.

Use Cases

  • Architectural Blueprinting: Designing scalable system architectures by selecting appropriate design patterns and defining component layers before implementation.
  • Code Reuse Analysis: Conducting thorough searches within existing codebases to identify reuse opportunities and prevent redundant development efforts.
  • Technical Documentation: Generating detailed implementation plans (plan.md) and research findings (research.md) to maintain project alignment and documentation standards.
  • API & Data Model Design: Defining structured API contracts and entity-relationship models early in the lifecycle to ensure data integrity and smooth integration.
  • Complexity Estimation: Predicting task counts and development duration to improve project management and resource allocation.
nameplanning-phase
description"Standard Operating Procedure for /plan phase. Covers research depth, code reuse detection, design pattern selection, and architecture planning."
allowed-toolsRead, Grep, Glob, Bash

Planning Phase: Quick Reference

Purpose: Generate implementation plan with reuse analysis from spec.md, ensuring alignment with project documentation and maximizing code reuse.

Phase Overview

Inputs:

  • specs/NNN-slug/spec.md - Feature specification
  • docs/project/*.md - 8 project documentation files (if available)

Outputs:

  • specs/NNN-slug/plan.md - Implementation plan
  • specs/NNN-slug/research.md - Research findings and reuse opportunities

Expected duration: 1-3 hours

Quick Start Checklist

Before you begin:

  • Specification phase completed (spec.md exists)
  • Understand feature requirements
  • Project documentation available (or brownfield codebase)

Core workflow:

  1. ✅ Load Project Documentation - Read all 8 docs for constraints
  2. ✅ Research Code Reuse - Search before designing
  3. ✅ Design Architecture - Components, layers, patterns
  4. ✅ Plan Data Model - Entities, relationships, migrations
  5. ✅ Define API Contracts - Endpoints, schemas, validation
  6. ✅ Plan Testing Strategy - Unit, integration, E2E coverage
  7. ✅ Estimate Complexity - Task count prediction

Detailed Resources

🎯 Core Workflow

📊 Planning Artifacts

🔧 Estimation & Validation

Completion Criteria

Required:

  • plan.md created with architecture, components, API contracts
  • research.md created with reuse findings and project context
  • All 8 project docs read (or brownfield research complete)
  • Code reuse opportunities identified (5-15 expected)
  • Testing strategy defined (unit, integration, E2E)
  • Complexity estimated (20-30 tasks predicted)

Optional (if applicable):

  • Data model ERD created (Mermaid diagram)
  • API contracts in OpenAPI format
  • UI/UX flow diagram (for UI features)

Next Phase

After planning complete: → /tasks - Break down into concrete implementation tasks

See also: