architecture

Design system components, interfaces, and data flow into a traceable ARCHITECTURE.md document.

3|Updated Oct 28, 2020
One-click install
npx skills add https://github.com/k0d3x8its/dotfiles --skill architecture-k0d3x8its
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: architecture
Source: https://github.com/k0d3x8its/dotfiles/tree/main/claude/.claude/skills/architecture
Command: npx skills add https://github.com/k0d3x8its/dotfiles --skill architecture-k0d3x8its

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often jump from requirements straight into code without a coherent system design, leaving components untraceable to the requirements they serve and design docs that drift out of sync. This Skill turns a requirements spec into a living architecture document where every component traces back to a specific FR/NFR. ## Core Features & Use Cases - Traceable system design: Generates docs/ARCHITECTURE.md with components, interfaces, data flow, and a bidirectional FR/NFR traceability table checked in both directions. - Living document revisions: Edits an existing architecture doc in place rather than appending, designing only gaps so unchanged components are not redesigned. - Built-in gates: Enforces an encryption check before writing, a hard gate against unjustified design decisions, and a conditional security review offer when the design exposes a network or sensitive-data surface. - Use Case: After running /requirements on a new feature, invoke /architecture to produce a mermaid system diagram, component list, and traceability table, then hand off to /write-plan for task breakdown. ## Quick Start Run /architecture with a short feature note after your docs/REQUIREMENTS.md exists to generate or revise the system design document.

Frequently Asked Questions about architecture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a software architecture document from requirements?▼

Run /architecture after producing docs/REQUIREMENTS.md. The Skill reads the spec, designs components and interfaces for uncovered FR/NFR items, and writes docs/ARCHITECTURE.md from a template with a full traceability table.

How to keep an architecture doc in sync with changing requirements?▼

Re-run /architecture whenever requirements change. It diffs the FR/NFR set against the existing traceability table, designs only what is new or changed, and edits the document in place so it stays a consistent snapshot of the current system.

Does the architecture skill work without a requirements document?▼

No. The Skill stops and recommends running /requirements first if docs/REQUIREMENTS.md is missing, because every component must trace to a specific FR or NFR and designing without a spec defeats traceability.

What is the difference between ARCHITECTURE.md and an ADR?▼

ARCHITECTURE.md is a living whole-system design updated in place as the system evolves. An ADR in docs/adr/ freezes one point-in-time decision with its rejected alternatives. Major tradeoffs worth preserving permanently belong in an ADR.

Why does the architecture skill refuse to write the design file?▼

It enforces an encryption gate: if .gitattributes lacks a git-crypt filter pattern for docs/ARCHITECTURE.md, it stops rather than writing a plaintext file you believe is encrypted. Run /encrypt first or explicitly confirm a plaintext exception.

When does the skill recommend a threat model review?▼

The offer fires only when the design names an HTTP or network-reachable component, or one handling auth, PII, payments, or credentials. It suggests /threat-model --design before /write-plan but never runs it automatically.