What it is
BMAD-METHOD (Breakthrough Method for Agile AI-Driven Development) is an open-source system for turning an idea or change request into working software without giving up the thinking. Small changes go straight to build; larger work gets a deeper planning loop.
I am not the author of BMAD. I contributed expansion-pack agent-system templates — role schemas, configuration, and the packing material you need if you want more than a single generic agent.
Why it matters
Agent repos rot into a pile of slightly different Markdown personas. Templates that agree on roles and config are how a method stays reusable. That is the slice I pushed: Part 1 Google Cloud setup, Part 2 agent system templates, Part 3 a configuration guide, as attributed in career-database from the upstream pull requests.
How it works
Upstream documents a delivery loop: clarify, plan, build, verify, then learn. My commits sit in the expansion-pack path, not in the core brand. I also kept a longer prompt-engineering note, AI-Context-Document, that I used with this work.
Status
The project later grew past 51k stars on GitHub. That number is a fact about the repo’s popularity, not a claim that my three pull requests caused it. Code: bmad-code-org/BMAD-METHOD.
Related
- AI-Context-Document — the prompt-engineering notes I used alongside that contribution