Skip to main content
The BMad Method includes a team of specialized AI agents, each with distinct roles, personalities, and capabilities. These agents collaborate throughout the software development lifecycle.

Agent Team

John - Product Manager

PRD creation, requirements discovery, stakeholder alignment

Winston - Architect

Distributed systems, cloud infrastructure, API design

Amelia - Developer

Story execution, test-driven development, code implementation

Quinn - QA Engineer

Test automation, API testing, E2E testing, coverage analysis

Bob - Scrum Master

Sprint planning, story preparation, agile ceremonies

Sally - UX Designer

User research, interaction design, UI patterns

Mary - Business Analyst

Market research, competitive analysis, requirements elicitation

Paige - Technical Writer

Documentation, Mermaid diagrams, standards compliance

Barry - Quick Flow Solo Dev

Rapid spec creation, lean implementation, minimum ceremony

Agent Collaboration

Agents work together across different phases:

Analysis Phase

  • Mary (Analyst) conducts market, domain, and technical research
  • Mary creates product briefs

Planning Phase

  • John (PM) creates and validates PRDs
  • Sally (UX Designer) designs user experiences

Solutioning Phase

  • Winston (Architect) creates technical architecture
  • John (PM) creates epics and stories
  • John, Winston, Sally verify implementation readiness

Implementation Phase

  • Bob (Scrum Master) plans sprints and prepares stories
  • Amelia (Developer) implements stories with TDD
  • Quinn (QA) generates automated tests
  • Amelia performs code reviews

Documentation Phase

  • Paige (Tech Writer) documents projects and explains concepts

Invoking Agents

Agents are invoked through their menu commands. Each agent has specific triggers:
# Examples
BP  # Mary - Brainstorm Project
CP  # John - Create PRD
CU  # Sally - Create UX Design
CA  # Winston - Create Architecture
SP  # Bob - Sprint Planning
DS  # Amelia - Dev Story
QA  # Quinn - Generate Tests
WD  # Paige - Write Document
See individual agent pages for complete command references.

Agent Configuration

All agents are defined in YAML files located at:
src/bmm/agents/
โ”œโ”€โ”€ pm.agent.yaml
โ”œโ”€โ”€ architect.agent.yaml
โ”œโ”€โ”€ dev.agent.yaml
โ”œโ”€โ”€ qa.agent.yaml
โ”œโ”€โ”€ sm.agent.yaml
โ”œโ”€โ”€ ux-designer.agent.yaml
โ”œโ”€โ”€ analyst.agent.yaml
โ””โ”€โ”€ tech-writer/
    โ””โ”€โ”€ tech-writer.agent.yaml
Each agent configuration includes:
  • metadata: ID, name, title, icon, capabilities
  • persona: Role, identity, communication style, principles
  • menu: Available commands and workflows
  • critical_actions: Key behaviors (for some agents)
  • prompts: Welcome messages and guidance (for some agents)

Build docs developers (and LLMs) love