bmad-architecture

Produces an architecture spine of invariants and decisions distilled from a memlog.

1|Updated Jun 2, 2026
One-click install
npx skills add https://github.com/SevWren/Daily-Motivation-Brain-Helper --skill bmad-architecture-sevwren
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-architecture
Source: https://github.com/SevWren/Daily-Motivation-Brain-Helper/tree/main/.claude/skills/bmad-architecture
Command: npx skills add https://github.com/SevWren/Daily-Motivation-Brain-Helper --skill bmad-architecture-sevwren

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Architecture documents often balloon into bloated prose or silently diverge when independent builders make incompatible choices. This Skill produces a lean architecture spine that fixes only the durable invariants — paradigm, boundaries, state mutation rules, shared data ownership — so everything built from it stays consistent. ## Core Features & Use Cases - Architecture Spine Generation: Distills decisions into stable AD-n blocks (Binds/Prevents/Rule) with mermaid diagrams, a pinned stack table, and an explicit Deferred section. - Coaching or Fast Path: Elicits decisions through open-ended coaching by default, or drafts the full spine with [ASSUMPTION] tags when speed is requested. - Memlog-Driven Workflow: Records every decision, constraint, and open question in an append-only .memlog.md, enabling resumable runs and stable decision IDs across updates. - Reviewer Gate: Runs a deterministic lint_spine.py pass plus parallel reviewer subagents before finalizing or validating a spine. - Use Case: Given a spec package or brownfield codebase, produce ARCHITECTURE-SPINE.md at initiative, feature, or epic altitude that downstream epic and story planning can cite. ## Quick Start Ask the assistant to create the architecture spine for your project from your spec or existing codebase, choosing the coaching path to work through the key decisions together.

Frequently Asked Questions about bmad-architecture

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

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

Invoke the skill with your spec package and it distills an architecture spine of invariants and AD-n decision blocks. It offers a coaching path that elicits decisions from you, or a fast path that drafts the spine with tagged assumptions for review.

What is an architecture spine versus a full architecture document?

A spine fixes only the invariants that keep independently built units consistent — paradigm, boundaries, state rules, shared data ownership. Structural details like stack and source tree are minimal seed content, and fuller prose documents are optional renderings produced only when the audience needs them.

Can this skill derive architecture from an existing codebase?

Yes. For brownfield projects it reads the existing code and project-context.md to ratify conventions already present rather than inventing new ones. Decisions already settled by reality are tagged as adopted instead of re-derived.

How does the skill validate an architecture spine?

The Reviewer Gate runs lint_spine.py for mechanical checks like duplicate AD IDs, missing Binds/Prevents/Rule fields, and unpinned stack versions, then dispatches parallel reviewer subagents against a good-spine checklist. Under the validate intent it delivers an HTML report without changing the spine.

Does the skill work without interactive user input?

Yes, headless mode infers intent, altitude, and inputs from the activation payload, records inferences as assumptions, and runs the full reviewer gate non-interactively. It ends with a JSON status report listing the spine path, open questions, and any conflicts.