Preview — full styling will appear after the next deploy completes.

bmad-method

Scale-Adaptive Routing

Quick Flow · BMad Method · Enterprise

Not every change needs the same depth of process. BMAD routes work to one of three tracks — Quick Flow, BMad Method, or Enterprise — based on scope, risk, and coordination requirements. The same routing logic also distinguishes greenfield from brownfield projects.

flowchart TD
    W([Work Item]) --> Q{Scope<br>Assessment}
    Q -->|small / well-understood| QF[Quick Flow<br>Brief context doc<br>then implement]
    Q -->|standard feature| BM[BMad Method<br>Full four-phase<br>artifact chain]
    Q -->|regulated / cross-team| EN[Enterprise<br>Full chain plus<br>compliance artifacts]
    QF --> BFQ{Greenfield<br>or Brownfield?}
    BM --> BFQ
    EN --> BFQ
    BFQ -->|new codebase| GF[Greenfield track]
    BFQ -->|existing codebase| BR[Brownfield track<br>add codebase survey]

A common failure mode for structured methodologies is applying heavyweight process uniformly. A two-line bug fix that requires a full PRD, user stories, and an architecture document isn't rigorous — it's wasteful. BMAD addresses this with scale-adaptive routing: a lightweight assessment at the start of any work item determines which track to follow.

Quick Flow is for small, well-understood changes: bug fixes, configuration updates, minor additions. It skips most of the artifact chain and goes straight to implementation with a brief context document. BMad Method is the standard track for new features and non-trivial work: the full four-phase cycle applies. Enterprise adds compliance documentation, governance checkpoints, and cross-team coordination artifacts for regulated or large-scale projects.

The routing logic also distinguishes greenfield projects (new codebase, no inherited constraints) from brownfield projects (existing codebase, technical debt, integration points). A brownfield Quick Flow still needs a context document describing the existing system boundaries. The same three tracks apply to both, but brownfield projects carry an additional input — a codebase survey — that travels alongside all other artifacts.