bmad-method

Coordinates a virtual agile team for structured, end-to-end AI-assisted software development workflows.

1|1|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/gaoqiongxie/skills-ai --skill bmad-method-gaoqiongxie
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-method
Source: https://github.com/gaoqiongxie/skills-ai/tree/main/bmad-method
Command: npx skills add https://github.com/gaoqiongxie/skills-ai --skill bmad-method-gaoqiongxie

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the common pain points of AI-assisted software development: single-turn chat context loss, forgotten architectural decisions, disorganized ad-hoc coding, and unmaintainable output when building software with AI.

Core Features & Use Cases

  • Multi-Role AI Collaboration: Simulates a full agile development team with specialized AI roles (analyst, product manager, architect, developer, QA, UX designer) to handle dedicated tasks and ensure output quality.
  • Dual Workflow Paths: Offers a quick path for small tasks like bug fixes and minor feature updates, and a full 7-stage path for complex projects like new product builds and system refactoring.
  • Context Optimization: Uses document sharding and context engineering to load only relevant information for each development task, avoiding context window overload and maintaining consistency with earlier project decisions. Typical use cases include building a new SaaS product from scratch, refactoring a legacy monolithic application, or fixing critical production bugs with full traceability to original requirements.

Quick Start

Use the bmad-method skill to guide you through building a complete user authentication module from requirement gathering to code review using the full AI dev team workflow.

Frequently Asked Questions about bmad-method

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

FAQPage Schema
How do I maintain architectural context and prevent context loss in AI-assisted software development?

To maintain architectural context in AI-assisted software development, use a multi-role virtual agile team framework to simulate dedicated analysts, architects, and developers, ensuring earlier project decisions remain consistent throughout the workflow.

What is the best way to manage complex new product builds with AI code generation?

The best way to manage complex new product builds with AI code generation is deploying a full 7-stage development path that handles structured requirement documentation, technical architecture design, task sharding, and standardized code review.

Can I use a virtual dev team workflow for quick bug fixes and minor feature iterations?

Yes, you can use a virtual dev team workflow for quick bug fixes and minor feature iterations by utilizing a dedicated quick path, which bypasses the full 7-stage process to resolve small tasks efficiently with traceability.

How does context engineering prevent AI context window overload during system refactoring?

Context engineering prevents AI context window overload during system refactoring by applying document sharding, which loads only the relevant information needed for each specific development task instead of the entire project scope.

Why does AI-generated code become unmaintainable during ad-hoc coding sessions?

AI-generated code becomes unmaintainable during ad-hoc coding sessions because of disorganized workflows and forgotten architectural decisions, which a structured multi-role collaboration framework solves by enforcing standardized code review and requirement documentation.