Tools: Ultimate Guide: BMAD-Method Workflows Deep Dive: From Idea to Production (Part 2)
The Four Development Phases
Phase 1: Analysis (Optional but Valuable)
Phase 2: Planning (Required)
Phase 3: Solutioning
Phase 4: Implementation
Quick Flow: The Fast Path
When Quick Flow Shines
Full Planning: Building Products Right
Complete Walkthrough: Building a Habit Tracker
Party Mode: Multi-Agent Collaboration
Enabling Party Mode
Party Mode Use Cases
Context Management: The Secret Sauce
Workflow Quick Reference
Tips for Effective Workflow Usage
1. Use Fresh Chats
2. Trust the Process (Initially)
3. Let Agents Guide You
4. Review Artifacts
5. Use Help When Stuck
Coming Next In Part 1, we introduced BMAD-Method and its philosophy of AI as collaborator. Now let's dive deep into the workflows that make it practical. BMAD organizes development into four distinct phases, each with specific goals and outputs: This is where ideas become clearer. You're not committing to anything yet—just exploring. Example: Brainstorming Session This is where requirements become concrete. The output differs based on your track: Quick Flow → Tech Spec Full BMad Method → PRD Now we design how to build it. This phase creates the technical foundation. Architecture Creation Breaking Down into Epics and Stories Implementation Readiness Check This is where code gets written. BMAD provides structure without micromanaging. The Development Cycle For each story, you follow this pattern: For smaller work, Quick Flow condenses everything into 3 commands: For new products or major features, the full path ensures nothing is missed. Step 1: Product Brief Step 4: Epics & Stories Step 5: Implementation One of BMAD's most powerful features is Party Mode—bringing multiple agents together for collaborative problem-solving. Troubleshooting Session Architecture Decision BMAD maintains context across sessions through file-based artifacts: Start a new chat for each major workflow. This keeps context focused and prevents confusion. Follow the recommended path until you understand it. Then customize. Agents ask questions for a reason. Answer thoughtfully—the quality of outputs depends on input quality. Before moving to the next phase, review what was generated. Catch issues early. Part 3: BMad Builder — Learn how to create custom agents tailored to your domain, build specialized workflows, and package them into shareable modules. Understanding workflows is the key to unlocking BMAD's full potential. Start with Quick Flow for small work, graduate to Full Planning for products, and use Party Mode when you need diverse perspectives. Questions about workflows? Ask in the comments! Templates let you quickly answer FAQs or store snippets for re-use. Are you sure you want to ? It will become hidden in your post, but will still be visible via the comment's permalink. as well , this person and/or
┌─────────────────────────────────────────────────────────────────┐ │ Phase 1: ANALYSIS │ Phase 2: PLANNING │ │ ───────────────── │ ─────────────────── │ │ • Brainstorming │ • PRD (Product Requirements) │ │ • Research │ • Tech Spec (Quick Flow) │ │ • Product Brief │ • User personas & journeys │ │ │ │ ├─────────────────────────┼───────────────────────────────────────┤ │ Phase 3: SOLUTIONING │ Phase 4: IMPLEMENTATION │ │ ───────────────────── │ ─────────────────────── │ │ • Architecture │ • Sprint planning │ │ • Epics & Stories │ • Story development │ │ • UX Design │ • Code review │ │ • Implementation Check │ • Retrospectives │ └─────────────────────────┴───────────────────────────────────────┘ CODE_BLOCK: ┌─────────────────────────────────────────────────────────────────┐ │ Phase 1: ANALYSIS │ Phase 2: PLANNING │ │ ───────────────── │ ─────────────────── │ │ • Brainstorming │ • PRD (Product Requirements) │ │ • Research │ • Tech Spec (Quick Flow) │ │ • Product Brief │ • User personas & journeys │ │ │ │ ├─────────────────────────┼───────────────────────────────────────┤ │ Phase 3: SOLUTIONING │ Phase 4: IMPLEMENTATION │ │ ───────────────────── │ ─────────────────────── │ │ • Architecture │ • Sprint planning │ │ • Epics & Stories │ • Story development │ │ • UX Design │ • Code review │ │ • Implementation Check │ • Retrospectives │ └─────────────────────────┴───────────────────────────────────────┘ CODE_BLOCK: ┌─────────────────────────────────────────────────────────────────┐ │ Phase 1: ANALYSIS │ Phase 2: PLANNING │ │ ───────────────── │ ─────────────────── │ │ • Brainstorming │ • PRD (Product Requirements) │ │ • Research │ • Tech Spec (Quick Flow) │ │ • Product Brief │ • User personas & journeys │ │ │ │ ├─────────────────────────┼───────────────────────────────────────┤ │ Phase 3: SOLUTIONING │ Phase 4: IMPLEMENTATION │ │ ───────────────────── │ ─────────────────────── │ │ • Architecture │ • Sprint planning │ │ • Epics & Stories │ • Story development │ │ • UX Design │ • Code review │ │ • Implementation Check │ • Retrospectives │ └─────────────────────────┴───────────────────────────────────────┘ CODE_BLOCK: You: /bmad-brainstorming Analyst: Welcome! I'm here to help you explore ideas systematically. What problem or opportunity are you thinking about? You: I want to build a tool that helps developers manage their AI prompts across projects. Analyst: Interesting! Let's explore this from multiple angles. 🎯 PROBLEM SPACE:
- Who experiences this pain most acutely?
- What do they currently do instead?
- What makes this urgent now? Would you like to start with user research or competitive analysis? CODE_BLOCK: You: /bmad-brainstorming Analyst: Welcome! I'm here to help you explore ideas systematically. What problem or opportunity are you thinking about? You: I want to build a tool that helps developers manage their AI prompts across projects. Analyst: Interesting! Let's explore this from multiple angles. 🎯 PROBLEM SPACE:
- Who experiences this pain most acutely?
- What do they currently do instead?
- What makes this urgent now? Would you like to start with user research or competitive analysis? CODE_BLOCK: You: /bmad-brainstorming Analyst: Welcome! I'm here to help you explore ideas systematically. What problem or opportunity are you thinking about? You: I want to build a tool that helps developers manage their AI prompts across projects. Analyst: Interesting! Let's explore this from multiple angles. 🎯 PROBLEM SPACE:
- Who experiences this pain most acutely?
- What do they currently do instead?
- What makes this urgent now? Would you like to start with user research or competitive analysis? CODE_BLOCK: You: /quick-spec BMAD: I'll analyze your project and create a technical specification. [Reads existing code, understands stack] 📄 Generated: _bmad-output/tech-spec.md Summary: - 8 stories identified
- Estimated complexity: Medium
- Recommended approach: Incremental implementation CODE_BLOCK: You: /quick-spec BMAD: I'll analyze your project and create a technical specification. [Reads existing code, understands stack] 📄 Generated: _bmad-output/tech-spec.md Summary: - 8 stories identified
- Estimated complexity: Medium
- Recommended approach: Incremental implementation CODE_BLOCK: You: /quick-spec BMAD: I'll analyze your project and create a technical specification. [Reads existing code, understands stack] 📄 Generated: _bmad-output/tech-spec.md Summary: - 8 stories identified
- Estimated complexity: Medium
- Recommended approach: Incremental implementation CODE_BLOCK: You: /create-prd PM: Let's build a comprehensive Product Requirements Document. I'll guide you through each section. First, let's define the problem statement. What problem does this product solve, and for whom? [Guides through: Problem, Users, Personas, Features, Success Metrics, Risks] 📄 Generated: _bmad-output/PRD.md CODE_BLOCK: You: /create-prd PM: Let's build a comprehensive Product Requirements Document. I'll guide you through each section. First, let's define the problem statement. What problem does this product solve, and for whom? [Guides through: Problem, Users, Personas, Features, Success Metrics, Risks] 📄 Generated: _bmad-output/PRD.md CODE_BLOCK: You: /create-prd PM: Let's build a comprehensive Product Requirements Document. I'll guide you through each section. First, let's define the problem statement. What problem does this product solve, and for whom? [Guides through: Problem, Users, Personas, Features, Success Metrics, Risks] 📄 Generated: _bmad-output/PRD.md CODE_BLOCK: You: /create-architecture Architect: I've reviewed your PRD. Let's design the system architecture. Based on your requirements, I'm considering:
- Microservices vs Monolith: Given scale requirements...
- Database: PostgreSQL for relational data, Redis for caching
- API: REST with OpenAPI specification Let's discuss the key architectural decisions. 📄 Generated: _bmad-output/architecture.md CODE_BLOCK: You: /create-architecture Architect: I've reviewed your PRD. Let's design the system architecture. Based on your requirements, I'm considering:
- Microservices vs Monolith: Given scale requirements...
- Database: PostgreSQL for relational data, Redis for caching
- API: REST with OpenAPI specification Let's discuss the key architectural decisions. 📄 Generated: _bmad-output/architecture.md CODE_BLOCK: You: /create-architecture Architect: I've reviewed your PRD. Let's design the system architecture. Based on your requirements, I'm considering:
- Microservices vs Monolith: Given scale requirements...
- Database: PostgreSQL for relational data, Redis for caching
- API: REST with OpenAPI specification Let's discuss the key architectural decisions. 📄 Generated: _bmad-output/architecture.md CODE_BLOCK: You: /create-epics-and-stories PM: I'll break down the PRD into implementable units. [Reads PRD and Architecture] 📁 Created Epic Structure: Epic 1: User Authentication (5 stories) Epic 2: Core API (8 stories) Epic 3: Data Management (4 stories) Epic 4: Admin Dashboard (6 stories) Total: 23 stories across 4 epics 📄 Generated: _bmad-output/epics/ CODE_BLOCK: You: /create-epics-and-stories PM: I'll break down the PRD into implementable units. [Reads PRD and Architecture] 📁 Created Epic Structure: Epic 1: User Authentication (5 stories) Epic 2: Core API (8 stories) Epic 3: Data Management (4 stories) Epic 4: Admin Dashboard (6 stories) Total: 23 stories across 4 epics 📄 Generated: _bmad-output/epics/ CODE_BLOCK: You: /create-epics-and-stories PM: I'll break down the PRD into implementable units. [Reads PRD and Architecture] 📁 Created Epic Structure: Epic 1: User Authentication (5 stories) Epic 2: Core API (8 stories) Epic 3: Data Management (4 stories) Epic 4: Admin Dashboard (6 stories) Total: 23 stories across 4 epics 📄 Generated: _bmad-output/epics/ CODE_BLOCK: You: /check-implementation-readiness Architect: I'll validate that all planning documents are cohesive and ready for implementation. Checking: ✅ PRD → Architecture alignment ✅ Architecture → Stories traceability ✅ Technical feasibility of all stories ⚠️ Story E2-S3 references undefined API endpoint Recommendation: Update Architecture section 4.2 before starting implementation. CODE_BLOCK: You: /check-implementation-readiness Architect: I'll validate that all planning documents are cohesive and ready for implementation. Checking: ✅ PRD → Architecture alignment ✅ Architecture → Stories traceability ✅ Technical feasibility of all stories ⚠️ Story E2-S3 references undefined API endpoint Recommendation: Update Architecture section 4.2 before starting implementation. CODE_BLOCK: You: /check-implementation-readiness Architect: I'll validate that all planning documents are cohesive and ready for implementation. Checking: ✅ PRD → Architecture alignment ✅ Architecture → Stories traceability ✅ Technical feasibility of all stories ⚠️ Story E2-S3 references undefined API endpoint Recommendation: Update Architecture section 4.2 before starting implementation. CODE_BLOCK: You: /sprint-planning SM: Let's set up sprint tracking. Based on your epics, I recommend: Sprint 1 (Foundation):
- E1-S1: Database schema
- E1-S2: User model
- E1-S3: Auth endpoints Sprint 2 (Core Features):
- E2-S1: API scaffolding
- E2-S2: CRUD operations ... 📄 Generated: _bmad-output/sprint-status.yaml CODE_BLOCK: You: /sprint-planning SM: Let's set up sprint tracking. Based on your epics, I recommend: Sprint 1 (Foundation):
- E1-S1: Database schema
- E1-S2: User model
- E1-S3: Auth endpoints Sprint 2 (Core Features):
- E2-S1: API scaffolding
- E2-S2: CRUD operations ... 📄 Generated: _bmad-output/sprint-status.yaml CODE_BLOCK: You: /sprint-planning SM: Let's set up sprint tracking. Based on your epics, I recommend: Sprint 1 (Foundation):
- E1-S1: Database schema
- E1-S2: User model
- E1-S3: Auth endpoints Sprint 2 (Core Features):
- E2-S1: API scaffolding
- E2-S2: CRUD operations ... 📄 Generated: _bmad-output/sprint-status.yaml COMMAND_BLOCK: # 1. Create detailed story file /create-story E1-S1 # 2. Implement the story /dev-story E1-S1 # 3. Review the implementation /code-review E1-S1 COMMAND_BLOCK: # 1. Create detailed story file /create-story E1-S1 # 2. Implement the story /dev-story E1-S1 # 3. Review the implementation /code-review E1-S1 COMMAND_BLOCK: # 1. Create detailed story file /create-story E1-S1 # 2. Implement the story /dev-story E1-S1 # 3. Review the implementation /code-review E1-S1 CODE_BLOCK: ┌────────────────────────────────────────────────────────┐ │ QUICK FLOW │ │ │ │ /quick-spec → /dev-story → /code-review │ │
↓
↓
↓ │ │ tech-spec.md implement validate │ │ with stories each story quality │ └────────────────────────────────────────────────────────┘ CODE_BLOCK: ┌────────────────────────────────────────────────────────┐ │ QUICK FLOW │ │ │ │ /quick-spec → /dev-story → /code-review │ │
↓
↓
↓ │ │ tech-spec.md implement validate │ │ with stories each story quality │ └────────────────────────────────────────────────────────┘ CODE_BLOCK: ┌────────────────────────────────────────────────────────┐ │ QUICK FLOW │ │ │ │ /quick-spec → /dev-story → /code-review │ │
↓
↓
↓ │ │ tech-spec.md implement validate │ │ with stories each story quality │ └────────────────────────────────────────────────────────┘ CODE_BLOCK: You: /quick-spec The login button doesn't work on mobile Safari. BMAD: [Analyzes codebase, identifies issue] Created tech-spec with 2 stories: 1. Fix Safari-specific CSS flexbox issue 2. Add mobile browser testing You: /dev-story 1 DEV: [Implements fix] Fixed: Updated button container to use -webkit-flex for Safari compatibility. You: /code-review DEV: ✅ Fix applied correctly ✅ No regression in other browsers ⚠️ Consider adding browser testing to CI CODE_BLOCK: You: /quick-spec The login button doesn't work on mobile Safari. BMAD: [Analyzes codebase, identifies issue] Created tech-spec with 2 stories: 1. Fix Safari-specific CSS flexbox issue 2. Add mobile browser testing You: /dev-story 1 DEV: [Implements fix] Fixed: Updated button container to use -webkit-flex for Safari compatibility. You: /code-review DEV: ✅ Fix applied correctly ✅ No regression in other browsers ⚠️ Consider adding browser testing to CI CODE_BLOCK: You: /quick-spec The login button doesn't work on mobile Safari. BMAD: [Analyzes codebase, identifies issue] Created tech-spec with 2 stories: 1. Fix Safari-specific CSS flexbox issue 2. Add mobile browser testing You: /dev-story 1 DEV: [Implements fix] Fixed: Updated button container to use -webkit-flex for Safari compatibility. You: /code-review DEV: ✅ Fix applied correctly ✅ No regression in other browsers ⚠️ Consider adding browser testing to CI CODE_BLOCK: You: /quick-spec Add dark mode toggle to settings page. BMAD: Created tech-spec with 4 stories: 1. Create theme context and provider 2. Add toggle component 3. Persist preference to localStorage 4. Update all components for theme awareness CODE_BLOCK: You: /quick-spec Add dark mode toggle to settings page. BMAD: Created tech-spec with 4 stories: 1. Create theme context and provider 2. Add toggle component 3. Persist preference to localStorage 4. Update all components for theme awareness CODE_BLOCK: You: /quick-spec Add dark mode toggle to settings page. BMAD: Created tech-spec with 4 stories: 1. Create theme context and provider 2. Add toggle component 3. Persist preference to localStorage 4. Update all components for theme awareness CODE_BLOCK: You: /create-product-brief Analyst: Let's capture the essence of your product. You: I want to build a habit tracker that uses gamification to keep users engaged. Analyst: Great! Let's define the core elements: Problem: Users struggle to maintain habits Solution: Gamified tracking with rewards Target Users: [Let's define this...] 📄 Generated: _bmad-output/product-brief.md CODE_BLOCK: You: /create-product-brief Analyst: Let's capture the essence of your product. You: I want to build a habit tracker that uses gamification to keep users engaged. Analyst: Great! Let's define the core elements: Problem: Users struggle to maintain habits Solution: Gamified tracking with rewards Target Users: [Let's define this...] 📄 Generated: _bmad-output/product-brief.md CODE_BLOCK: You: /create-product-brief Analyst: Let's capture the essence of your product. You: I want to build a habit tracker that uses gamification to keep users engaged. Analyst: Great! Let's define the core elements: Problem: Users struggle to maintain habits Solution: Gamified tracking with rewards Target Users: [Let's define this...] 📄 Generated: _bmad-output/product-brief.md CODE_BLOCK: You: /create-prd PM: Building on your product brief, let's create comprehensive requirements. [Guides through 12 sections including:]
- User Personas
- Feature Requirements (MoSCoW prioritized)
- User Journeys
- Success Metrics
- Technical Constraints
- Risk Analysis 📄 Generated: _bmad-output/PRD.md (comprehensive 15-page doc) CODE_BLOCK: You: /create-prd PM: Building on your product brief, let's create comprehensive requirements. [Guides through 12 sections including:]
- User Personas
- Feature Requirements (MoSCoW prioritized)
- User Journeys
- Success Metrics
- Technical Constraints
- Risk Analysis 📄 Generated: _bmad-output/PRD.md (comprehensive 15-page doc) CODE_BLOCK: You: /create-prd PM: Building on your product brief, let's create comprehensive requirements. [Guides through 12 sections including:]
- User Personas
- Feature Requirements (MoSCoW prioritized)
- User Journeys
- Success Metrics
- Technical Constraints
- Risk Analysis 📄 Generated: _bmad-output/PRD.md (comprehensive 15-page doc) CODE_BLOCK: You: /create-architecture Architect: I've reviewed the PRD. Here's my recommended architecture: Frontend: React Native (cross-platform requirement) Backend: Node.js + Express Database: PostgreSQL + Redis Key Decisions: 1. Real-time sync via WebSockets 2. Offline-first with local SQLite 3. Push notifications via Firebase Let's discuss the gamification engine design... 📄 Generated: _bmad-output/architecture.md CODE_BLOCK: You: /create-architecture Architect: I've reviewed the PRD. Here's my recommended architecture: Frontend: React Native (cross-platform requirement) Backend: Node.js + Express Database: PostgreSQL + Redis Key Decisions: 1. Real-time sync via WebSockets 2. Offline-first with local SQLite 3. Push notifications via Firebase Let's discuss the gamification engine design... 📄 Generated: _bmad-output/architecture.md CODE_BLOCK: You: /create-architecture Architect: I've reviewed the PRD. Here's my recommended architecture: Frontend: React Native (cross-platform requirement) Backend: Node.js + Express Database: PostgreSQL + Redis Key Decisions: 1. Real-time sync via WebSockets 2. Offline-first with local SQLite 3. Push notifications via Firebase Let's discuss the gamification engine design... 📄 Generated: _bmad-output/architecture.md CODE_BLOCK: You: /create-epics-and-stories PM: Breaking down into implementable work: Epic 1: User Onboarding (4 stories) Epic 2: Habit Management (6 stories) Epic 3: Gamification Engine (8 stories) Epic 4: Social Features (5 stories) Epic 5: Notifications (4 stories) Epic 6: Analytics Dashboard (3 stories) Total: 30 stories, estimated 6 sprints 📄 Generated: _bmad-output/epics/*.md CODE_BLOCK: You: /create-epics-and-stories PM: Breaking down into implementable work: Epic 1: User Onboarding (4 stories) Epic 2: Habit Management (6 stories) Epic 3: Gamification Engine (8 stories) Epic 4: Social Features (5 stories) Epic 5: Notifications (4 stories) Epic 6: Analytics Dashboard (3 stories) Total: 30 stories, estimated 6 sprints 📄 Generated: _bmad-output/epics/*.md CODE_BLOCK: You: /create-epics-and-stories PM: Breaking down into implementable work: Epic 1: User Onboarding (4 stories) Epic 2: Habit Management (6 stories) Epic 3: Gamification Engine (8 stories) Epic 4: Social Features (5 stories) Epic 5: Notifications (4 stories) Epic 6: Analytics Dashboard (3 stories) Total: 30 stories, estimated 6 sprints 📄 Generated: _bmad-output/epics/*.md COMMAND_BLOCK: You: /sprint-planning You: /create-story E1-S1 You: /dev-story E1-S1 You: /code-review ... [repeat for each story] You: /retrospective # After each epic COMMAND_BLOCK: You: /sprint-planning You: /create-story E1-S1 You: /dev-story E1-S1 You: /code-review ... [repeat for each story] You: /retrospective # After each epic COMMAND_BLOCK: You: /sprint-planning You: /create-story E1-S1 You: /dev-story E1-S1 You: /code-review ... [repeat for each story] You: /retrospective # After each epic CODE_BLOCK: You: /party-mode PM, Architect, Security [Party Mode Activated] PM: 👋 Product Manager here. Ready to ensure we're building the right thing. Architect: 🏗️ Architect present. Let's make sure it's built right. Security: 🔐 Security checking in. I'll keep us safe. You: We're adding a payment system. What should we consider? PM: From a product perspective:
- What payment methods do users expect?
- How does pricing affect conversion?
- What's our refund policy? Architect: Technically, I recommend:
- Use Stripe for PCI compliance
- Implement idempotency for reliability
- Consider webhook handling architecture Security: Critical considerations:
- Never store raw card numbers
- Implement rate limiting on payment endpoints
- Audit logging for all transactions
- Consider fraud detection integration CODE_BLOCK: You: /party-mode PM, Architect, Security [Party Mode Activated] PM: 👋 Product Manager here. Ready to ensure we're building the right thing. Architect: 🏗️ Architect present. Let's make sure it's built right. Security: 🔐 Security checking in. I'll keep us safe. You: We're adding a payment system. What should we consider? PM: From a product perspective:
- What payment methods do users expect?
- How does pricing affect conversion?
- What's our refund policy? Architect: Technically, I recommend:
- Use Stripe for PCI compliance
- Implement idempotency for reliability
- Consider webhook handling architecture Security: Critical considerations:
- Never store raw card numbers
- Implement rate limiting on payment endpoints
- Audit logging for all transactions
- Consider fraud detection integration CODE_BLOCK: You: /party-mode PM, Architect, Security [Party Mode Activated] PM: 👋 Product Manager here. Ready to ensure we're building the right thing. Architect: 🏗️ Architect present. Let's make sure it's built right. Security: 🔐 Security checking in. I'll keep us safe. You: We're adding a payment system. What should we consider? PM: From a product perspective:
- What payment methods do users expect?
- How does pricing affect conversion?
- What's our refund policy? Architect: Technically, I recommend:
- Use Stripe for PCI compliance
- Implement idempotency for reliability
- Consider webhook handling architecture Security: Critical considerations:
- Never store raw card numbers
- Implement rate limiting on payment endpoints
- Audit logging for all transactions
- Consider fraud detection integration CODE_BLOCK: /party-mode Architect, UX, Developer You: Let's review the dashboard design. UX: The layout prioritizes key metrics, but we should consider mobile responsiveness for field users. Architect: The real-time data requirements mean we'll need WebSocket connections—let's discuss connection pooling. Developer: I can implement this with React Query for optimistic updates. Estimated 2-3 days. CODE_BLOCK: /party-mode Architect, UX, Developer You: Let's review the dashboard design. UX: The layout prioritizes key metrics, but we should consider mobile responsiveness for field users. Architect: The real-time data requirements mean we'll need WebSocket connections—let's discuss connection pooling. Developer: I can implement this with React Query for optimistic updates. Estimated 2-3 days. CODE_BLOCK: /party-mode Architect, UX, Developer You: Let's review the dashboard design. UX: The layout prioritizes key metrics, but we should consider mobile responsiveness for field users. Architect: The real-time data requirements mean we'll need WebSocket connections—let's discuss connection pooling. Developer: I can implement this with React Query for optimistic updates. Estimated 2-3 days. CODE_BLOCK: /party-mode Developer, DevOps, DBA You: Production is slow. Query times jumped from 50ms to 2s. Developer: Let me check the recent code changes... The new feature adds a JOIN on the orders table. DBA: That table has 10M rows without proper indexing. Adding a composite index should help significantly. DevOps: I'm also seeing memory pressure on the DB instance. Consider scaling vertically or adding read replicas. CODE_BLOCK: /party-mode Developer, DevOps, DBA You: Production is slow. Query times jumped from 50ms to 2s. Developer: Let me check the recent code changes... The new feature adds a JOIN on the orders table. DBA: That table has 10M rows without proper indexing. Adding a composite index should help significantly. DevOps: I'm also seeing memory pressure on the DB instance. Consider scaling vertically or adding read replicas. CODE_BLOCK: /party-mode Developer, DevOps, DBA You: Production is slow. Query times jumped from 50ms to 2s. Developer: Let me check the recent code changes... The new feature adds a JOIN on the orders table. DBA: That table has 10M rows without proper indexing. Adding a composite index should help significantly. DevOps: I'm also seeing memory pressure on the DB instance. Consider scaling vertically or adding read replicas. CODE_BLOCK: /party-mode Architect, Security, PM You: Should we use microservices or monolith? PM: Our timeline is 3 months to MVP. How does this affect delivery speed? Architect: Monolith is faster for MVP, but microservices scale better. Given our user projections... Security: Microservices add attack surface but improve isolation. For healthcare data, I'd recommend at minimum separating the auth service. CODE_BLOCK: /party-mode Architect, Security, PM You: Should we use microservices or monolith? PM: Our timeline is 3 months to MVP. How does this affect delivery speed? Architect: Monolith is faster for MVP, but microservices scale better. Given our user projections... Security: Microservices add attack surface but improve isolation. For healthcare data, I'd recommend at minimum separating the auth service. CODE_BLOCK: /party-mode Architect, Security, PM You: Should we use microservices or monolith? PM: Our timeline is 3 months to MVP. How does this affect delivery speed? Architect: Monolith is faster for MVP, but microservices scale better. Given our user projections... Security: Microservices add attack surface but improve isolation. For healthcare data, I'd recommend at minimum separating the auth service. COMMAND_BLOCK: your-project/ ├── _bmad/ # BMAD configuration │ ├── agents/ # Agent definitions │ ├── workflows/ # Workflow configurations │ └── core-config.yaml # Project settings │ └── _bmad-output/ # Your artifacts ├── product-brief.md # Phase 1 output ├── PRD.md # Phase 2 output ├── architecture.md # Phase 3 output ├── epics/ # Story breakdowns │ ├── epic-1.md │ └── epic-2.md └── sprint-status.yaml # Current progress COMMAND_BLOCK: your-project/ ├── _bmad/ # BMAD configuration │ ├── agents/ # Agent definitions │ ├── workflows/ # Workflow configurations │ └── core-config.yaml # Project settings │ └── _bmad-output/ # Your artifacts ├── product-brief.md # Phase 1 output ├── PRD.md # Phase 2 output ├── architecture.md # Phase 3 output ├── epics/ # Story breakdowns │ ├── epic-1.md │ └── epic-2.md └── sprint-status.yaml # Current progress COMMAND_BLOCK: your-project/ ├── _bmad/ # BMAD configuration │ ├── agents/ # Agent definitions │ ├── workflows/ # Workflow configurations │ └── core-config.yaml # Project settings │ └── _bmad-output/ # Your artifacts ├── product-brief.md # Phase 1 output ├── PRD.md # Phase 2 output ├── architecture.md # Phase 3 output ├── epics/ # Story breakdowns │ ├── epic-1.md │ └── epic-2.md └── sprint-status.yaml # Current progress CODE_BLOCK: /bmad-help /bmad-help I just finished the PRD, what's next? /bmad-help How do I handle scope changes mid-sprint? CODE_BLOCK: /bmad-help /bmad-help I just finished the PRD, what's next? /bmad-help How do I handle scope changes mid-sprint? CODE_BLOCK: /bmad-help /bmad-help I just finished the PRD, what's next? /bmad-help How do I handle scope changes mid-sprint? - /bmad-brainstorming — Structured ideation with the Analyst agent - /bmad-bmm-research — Market analysis, competitive research, technical feasibility - /create-product-brief — Capture the essence before diving into details
- No context loss — Start a new chat, agents pick up where you left off
- Version control — Your planning docs are in git
- Team visibility — Everyone sees the same artifacts
- AI grounding — Agents reference concrete documents, not hallucinations