bmad-code-org/BMAD-METHOD
Developer ToolsA structured AI-assisted development method that turns ideas and change requests into planned, reviewed software.
What it does
BMad Method is an agile, AI-driven development workflow for teams that already use coding assistants but want to avoid letting unstated assumptions become code. It covers the full delivery loop: what to build, how the parts fit together, and how the design changes as you learn. The process scales to the work — small changes can go straight to implementation, while larger efforts get planning, architecture and testing input from specialized agent roles.
The method is delivered as installable skills and workflows for AI coding tools, with a Python-backed installer and scripts. It supports both greenfield prototypes and existing codebases, and its artifacts (briefs, specifications, architecture notes) can be carried into an existing delivery process rather than requiring an end-to-end switch.
Quick start
The README presents two release lines. The stable line, v6.12, requires Node.js 20.12+ and Python 3.10+, plus uv. Run this in your project:
npx bmad-method install
Then open the project in your AI coding tool, invoke bmad-build with the change you want, and call bmad-help for guidance on the next step. The v6.12 source and docs live on the V6.12 branch.
The v7 preview alpha targets coding tools that support skills and requires Node.js, npm and Git:
npx skills add bmad-code-org/BMAD-METHOD/skills/core-tools --skill '*'
npx skills add bmad-code-org/BMAD-METHOD/skills/method --skill '*'
Each command installs one module and prompts for the target coding tool. Core tools provide the bmad skill for setup and help, so they are required. Optional modules such as Toolsmith and Founder install the same way. After installation, ask the bmad skill to run bmad setup, then invoke bmad-build; bmad status checks versions and suggests what to run next.
A first end-to-end walkthrough is available in Build Your First Change, with separate guidance for existing codebases and choosing how much planning a change needs.
Constraints and compatibility
The stable line is frozen apart from bug fixes; the v7 preview is alpha and the README explicitly warns of breaking changes between updates, with a release ETA of mid-November 2026 that may move earlier. V7 requires an AI coding tool that understands skills, plus Git. Both routes require uv for setup and Python scripts, and the stable installer needs Node.js 20.12+ and Python 3.10+.
The project is MIT-licensed, with BMad and BMAD-METHOD names protected under a separate trademark policy. The README states that workflows and community access are not paywalled; optional support channels include a coffee donation link and corporate sponsorship contact.
Ecosystem
Beyond the core method, official modules cover skill and agent building, creative exploration, enterprise test architecture, unattended epic build/verify/retro loops, and game development across Unity, Unreal, Godot and Phaser. Teams can start with the core method and add modules only where the work calls for them.
HotRepo's take
The repository carries no recorded all-language daily rank, but its scale is hard to ignore: roughly 55,000 stars and 6,000 forks, and a Python-language daily snapshot on 2026-11-14 shows 21 stars gained and a 15th-place position on that language list. Because that is a single daily observation, it supports the fact that the project was active on that date, not a trend claim. The more interesting signal is structural: the README is splitting a frozen stable line from an alpha v7 that is still changing, which means readers should decide by stability needs first and pick the install path accordingly.
Written by AI from the project's README, updated . The project's own docs are the source of truth.
Hacker News
You might also like
- Fission-AI/OpenSpecA lightweight spec layer that keeps AI coding assistants and developers aligned before code is written.
- open-gsd/gsd-coreA phased workflow system that keeps AI coding agents context-lean across research, planning, execution, verification and shipping.
- google/elephant-goldfish-modelAn open-source plugin that helps coding agents draft software specifications and runs fresh-context reviewers to expose gaps before implementation.
- obra/superpowersA skill-based methodology that pushes coding agents to spec, plan, and review before writing code.