plan-architecture

Create a technical architecture specification.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "plan-architecture" with this command: npx skills add dthompson-jti/safety-check-app-concept/dthompson-jti-safety-check-app-concept-plan-architecture

Plan Architecture

Create a technical architecture specification.

When to Use

  • After PRD approval for large features

  • Complex technical decisions

  • Cross-cutting system changes

Artifacts

  • implementation_plan.md artifact

  • task.md updated via task_boundary

Approach

Phase 1: Requirement Mapping

  • Map PRD requirements to architectural components

  • Verify constraints (existing patterns, dependencies)

Phase 2: Options Evaluation

Generate 3-4 architectural options.

Consult expert frameworks:

  • SOLID Principles: Component design

  • CAP Theorem: Data consistency (if applicable)

  • OWASP Top 10: Security considerations

Use Socratic Debate for evaluation.

Phase 3: Detailed Specification

Data Flow Diagram

Show how data moves through the system.

Data Model

Define types, interfaces, atoms/state.

Component Hierarchy

ParentComponent ├── ChildA │ └── GrandchildA1 └── ChildB

State Management

Define atoms, derived state, effects.

File Manifest

Status File Changes

[NEW] src/components/Feature.tsx

New component

[MODIFY] src/atoms/index.ts

Add new atom

Phase 4: Risk Analysis

Reflexion as "Chaotic Junior Developer":

  • Attack Vector 1: How could this fail under rapid state changes?

  • Attack Vector 2: What breaks if API returns malformed data?

  • Attack Vector 3: Race conditions?

Phase 5: Phased Implementation

Break into atomic, verifiable phases.

Constraints

  • No code generation — architecture only

  • No one-off patterns — leverage existing documented patterns

  • Request user approval before BUILD phase

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

implement-feature

No summary provided by upstream source.

Repository SourceNeeds Review
General

explore-concept

No summary provided by upstream source.

Repository SourceNeeds Review
General

explore-ui-design

No summary provided by upstream source.

Repository SourceNeeds Review
General

plan-prd

No summary provided by upstream source.

Repository SourceNeeds Review