bmad-method 4.12.0 → 4.14.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (41) hide show
  1. package/CHANGELOG.md +14 -0
  2. package/CONTRIBUTING.md +128 -1
  3. package/GUIDING-PRINCIPLES.md +85 -0
  4. package/README.md +56 -6
  5. package/bmad-core/agents/analyst.md +3 -1
  6. package/bmad-core/agents/dev.md +3 -2
  7. package/bmad-core/agents/qa.md +15 -14
  8. package/bmad-core/data/bmad-kb.md +280 -19
  9. package/bmad-core/tasks/document-project.md +250 -322
  10. package/bmad-core/tasks/review-story.md +135 -0
  11. package/bmad-core/templates/story-tmpl.md +8 -0
  12. package/bmad-core/workflows/brownfield-fullstack.yml +36 -1
  13. package/bmad-core/workflows/brownfield-service.yml +36 -1
  14. package/bmad-core/workflows/brownfield-ui.yml +36 -1
  15. package/bmad-core/workflows/greenfield-fullstack.yml +36 -1
  16. package/bmad-core/workflows/greenfield-service.yml +36 -1
  17. package/bmad-core/workflows/greenfield-ui.yml +36 -1
  18. package/dist/agents/analyst.txt +227 -17
  19. package/dist/agents/bmad-master.txt +260 -24
  20. package/dist/agents/bmad-orchestrator.txt +227 -17
  21. package/dist/agents/dev.txt +6 -4
  22. package/dist/agents/pm.txt +25 -7
  23. package/dist/agents/po.txt +33 -7
  24. package/dist/agents/qa.txt +153 -14
  25. package/dist/agents/sm.txt +8 -0
  26. package/dist/expansion-packs/expansion-creator/agents/bmad-the-creator.txt +4 -12
  27. package/dist/teams/team-all.txt +635 -48
  28. package/dist/teams/team-fullstack.txt +476 -30
  29. package/dist/teams/team-ide-minimal.txt +419 -42
  30. package/dist/teams/team-no-ui.txt +332 -26
  31. package/docs/expansion-pack-ideas.md +121 -0
  32. package/docs/expansion-packs.md +265 -0
  33. package/docs/working-in-the-brownfield.md +362 -0
  34. package/expansion-packs/expansion-creator/tasks/create-agent.md +1 -1
  35. package/expansion-packs/expansion-creator/tasks/generate-expansion-pack.md +3 -11
  36. package/package.json +1 -1
  37. package/tools/installer/bin/bmad.js +12 -11
  38. package/tools/installer/config/install.config.yml +8 -3
  39. package/tools/installer/lib/ide-setup.js +71 -0
  40. package/tools/installer/package.json +1 -1
  41. package/tools/upgraders/v3-to-v4-upgrader.js +1 -0
@@ -0,0 +1,265 @@
1
+ # The Power of BMAD Expansion Packs
2
+
3
+ ## Overview
4
+
5
+ BMAD Method's expansion packs unlock the framework's true potential by extending its natural language AI orchestration to ANY domain. While the core framework focuses on software development, expansion packs transform BMAD into a universal AI agent system.
6
+
7
+ ## Why Expansion Packs?
8
+
9
+ ### Keep Core Lean
10
+
11
+ The core BMAD framework maintains its focus on software development, ensuring dev agents have maximum context for coding. Expansion packs handle everything else.
12
+
13
+ ### Domain Expertise
14
+
15
+ Each expansion pack provides deep, specialized knowledge without bloating the core system. Install only what you need.
16
+
17
+ ### Community Innovation
18
+
19
+ Anyone can create and share expansion packs, fostering a ecosystem of AI-powered solutions across all industries and interests.
20
+
21
+ ## Technical Expansion Packs
22
+
23
+ ### Game Development Pack
24
+
25
+ Transform your AI into a complete game development studio:
26
+
27
+ - **Game Designer**: Mechanics, balance, progression systems
28
+ - **Level Designer**: Map layouts, puzzle design, difficulty curves
29
+ - **Narrative Designer**: Story arcs, dialog trees, lore creation
30
+ - **Art Director**: Visual style guides, asset specifications
31
+ - **Sound Designer**: Audio direction, music themes, SFX planning
32
+
33
+ ### Mobile Development Pack
34
+
35
+ Specialized agents for mobile app creation:
36
+
37
+ - **iOS Specialist**: Swift/SwiftUI patterns, Apple guidelines
38
+ - **Android Expert**: Kotlin best practices, Material Design
39
+ - **Mobile UX Designer**: Touch interfaces, gesture patterns
40
+ - **App Store Optimizer**: ASO strategies, listing optimization
41
+ - **Performance Tuner**: Battery optimization, network efficiency
42
+
43
+ ### DevOps/Infrastructure Pack
44
+
45
+ Complete infrastructure automation team:
46
+
47
+ - **Cloud Architect**: AWS/Azure/GCP design patterns
48
+ - **Security Specialist**: Zero-trust implementation, compliance
49
+ - **SRE Expert**: Monitoring, alerting, incident response
50
+ - **Container Orchestrator**: Kubernetes, Docker optimization
51
+ - **Cost Optimizer**: Cloud spend analysis, resource right-sizing
52
+
53
+ ### Data Science Pack
54
+
55
+ AI-powered data analysis team:
56
+
57
+ - **Data Scientist**: Statistical analysis, ML model selection
58
+ - **Data Engineer**: Pipeline design, ETL processes
59
+ - **ML Engineer**: Model deployment, A/B testing
60
+ - **Visualization Expert**: Dashboard design, insight communication
61
+ - **Ethics Advisor**: Bias detection, fairness assessment
62
+
63
+ ## Non-Technical Expansion Packs
64
+
65
+ ### Business Strategy Pack
66
+
67
+ Complete business advisory team:
68
+
69
+ - **Strategy Consultant**: Market positioning, competitive analysis
70
+ - **Financial Analyst**: Projections, unit economics, funding strategies
71
+ - **Operations Manager**: Process optimization, efficiency improvements
72
+ - **Marketing Strategist**: Go-to-market plans, growth hacking
73
+ - **HR Advisor**: Talent strategies, culture building
74
+
75
+ ### Creative Writing Pack
76
+
77
+ Your personal writing team:
78
+
79
+ - **Plot Architect**: Three-act structure, story beats, pacing
80
+ - **Character Psychologist**: Deep motivations, authentic dialog
81
+ - **World Builder**: Consistent universes, cultural systems
82
+ - **Editor**: Style consistency, grammar, flow
83
+ - **Beta Reader**: Feedback simulation, plot hole detection
84
+
85
+ ### Health & Wellness Pack
86
+
87
+ Personal wellness coaching system:
88
+
89
+ - **Fitness Trainer**: Progressive overload, form correction
90
+ - **Nutritionist**: Macro planning, supplement guidance
91
+ - **Sleep Coach**: Circadian optimization, sleep hygiene
92
+ - **Stress Manager**: Coping strategies, work-life balance
93
+ - **Habit Engineer**: Behavior change, accountability systems
94
+
95
+ ### Education Pack
96
+
97
+ Complete learning design system:
98
+
99
+ - **Curriculum Architect**: Learning objectives, scope & sequence
100
+ - **Instructional Designer**: Engagement strategies, multimedia learning
101
+ - **Assessment Specialist**: Rubrics, formative/summative evaluation
102
+ - **Differentiation Expert**: Adaptive learning, special needs
103
+ - **EdTech Integrator**: Tool selection, digital pedagogy
104
+
105
+ ### Mental Health Support Pack
106
+
107
+ Therapeutic support system:
108
+
109
+ - **CBT Guide**: Cognitive restructuring, thought challenging
110
+ - **Mindfulness Teacher**: Meditation scripts, awareness exercises
111
+ - **Journal Therapist**: Reflective prompts, emotional processing
112
+ - **Crisis Support**: Coping strategies, safety planning
113
+ - **Habit Tracker**: Mood monitoring, trigger identification
114
+
115
+ ### Legal Assistant Pack
116
+
117
+ Legal document and research support:
118
+
119
+ - **Contract Analyst**: Term review, risk assessment
120
+ - **Legal Researcher**: Case law, precedent analysis
121
+ - **Document Drafter**: Template customization, clause libraries
122
+ - **Compliance Checker**: Regulatory alignment, audit prep
123
+ - **IP Advisor**: Patent strategies, trademark guidance
124
+
125
+ ### Real Estate Pack
126
+
127
+ Property investment and management:
128
+
129
+ - **Market Analyst**: Comparable analysis, trend prediction
130
+ - **Investment Calculator**: ROI modeling, cash flow analysis
131
+ - **Property Manager**: Tenant screening, maintenance scheduling
132
+ - **Flip Strategist**: Renovation ROI, project planning
133
+ - **Agent Assistant**: Listing optimization, showing prep
134
+
135
+ ## Unique & Innovative Packs
136
+
137
+ ### Role-Playing Game Master Pack
138
+
139
+ AI-powered tabletop RPG assistance:
140
+
141
+ - **World Master**: Dynamic world generation, NPC creation
142
+ - **Combat Referee**: Initiative tracking, rule clarification
143
+ - **Story Weaver**: Plot hooks, side quests, consequences
144
+ - **Character Builder**: Backstory generation, stat optimization
145
+ - **Loot Master**: Treasure generation, magic item creation
146
+
147
+ ### Life Event Planning Pack
148
+
149
+ Major life event coordination:
150
+
151
+ - **Wedding Planner**: Vendor coordination, timeline creation
152
+ - **Event Designer**: Theme development, decoration plans
153
+ - **Budget Manager**: Cost tracking, vendor negotiation
154
+ - **Guest Coordinator**: RSVP tracking, seating arrangements
155
+ - **Timeline Keeper**: Day-of scheduling, contingency planning
156
+
157
+ ### Hobby Mastery Pack
158
+
159
+ Deep dive into specific hobbies:
160
+
161
+ - **Garden Designer**: Plant selection, seasonal planning
162
+ - **Brew Master**: Recipe formulation, process optimization
163
+ - **Maker Assistant**: 3D printing, woodworking, crafts
164
+ - **Collection Curator**: Organization, valuation, trading
165
+ - **Photography Coach**: Composition, lighting, post-processing
166
+
167
+ ### Scientific Research Pack
168
+
169
+ Research acceleration tools:
170
+
171
+ - **Literature Reviewer**: Paper summarization, gap analysis
172
+ - **Hypothesis Generator**: Research question formulation
173
+ - **Methodology Designer**: Experiment planning, control design
174
+ - **Statistical Advisor**: Test selection, power analysis
175
+ - **Grant Writer**: Proposal structure, impact statements
176
+
177
+ ## Creating Your Own Expansion Pack
178
+
179
+ ### Step 1: Define Your Domain
180
+
181
+ What expertise are you capturing? What problems will it solve?
182
+
183
+ ### Step 2: Design Your Agents
184
+
185
+ Each agent should have:
186
+
187
+ - Clear expertise area
188
+ - Specific personality traits
189
+ - Defined capabilities
190
+ - Knowledge boundaries
191
+
192
+ ### Step 3: Create Tasks
193
+
194
+ Tasks should be:
195
+
196
+ - Step-by-step procedures
197
+ - Reusable across scenarios
198
+ - Clear and actionable
199
+ - Domain-specific
200
+
201
+ ### Step 4: Build Templates
202
+
203
+ Templates need:
204
+
205
+ - Structured output format
206
+ - Embedded LLM instructions
207
+ - Placeholders for customization
208
+ - Professional formatting
209
+
210
+ ### Step 5: Test & Iterate
211
+
212
+ - Use with real scenarios
213
+ - Gather user feedback
214
+ - Refine agent responses
215
+ - Improve task clarity
216
+
217
+ ### Step 6: Package & Share
218
+
219
+ - Create clear documentation
220
+ - Include usage examples
221
+ - Add to expansion-packs directory
222
+ - Share with community
223
+
224
+ ## The Future of Expansion Packs
225
+
226
+ ### Marketplace Potential
227
+
228
+ Imagine a future where:
229
+
230
+ - Professional expansion packs are sold
231
+ - Certified packs for regulated industries
232
+ - Community ratings and reviews
233
+ - Automatic updates and improvements
234
+
235
+ ### AI Agent Ecosystems
236
+
237
+ Expansion packs could enable:
238
+
239
+ - Cross-pack agent collaboration
240
+ - Industry-standard agent protocols
241
+ - Interoperable AI workflows
242
+ - Universal agent languages
243
+
244
+ ### Democratizing Expertise
245
+
246
+ Every expansion pack:
247
+
248
+ - Makes expert knowledge accessible
249
+ - Reduces barriers to entry
250
+ - Enables solo entrepreneurs
251
+ - Empowers small teams
252
+
253
+ ## Getting Started
254
+
255
+ 1. **Browse existing packs**: Check `expansion-packs/` directory
256
+ 2. **Install what you need**: Use the installer's expansion pack option
257
+ 3. **Create your own**: Use the expansion-creator pack
258
+ 4. **Share with others**: Submit PRs with new packs
259
+ 5. **Build the future**: Help shape AI-assisted work
260
+
261
+ ## Remember
262
+
263
+ The BMAD Method is more than a development framework - it's a platform for structuring human expertise into AI-accessible formats. Every expansion pack you create makes specialized knowledge more accessible to everyone.
264
+
265
+ **What expertise will you share with the world?**
@@ -0,0 +1,362 @@
1
+ # Working in the Brownfield: A Complete Guide
2
+
3
+ > 🚀 **HIGHLY RECOMMENDED: Use Gemini Web for Brownfield Documentation!**
4
+ >
5
+ > Gemini Web's 1M+ token context window can analyze your ENTIRE codebase at once:
6
+ >
7
+ > - Upload via GitHub URL
8
+ > - Upload up to 1000 files
9
+ > - Upload zipped project
10
+ >
11
+ > This is MUCH more cost-effective than IDE analysis which reads files one by one!
12
+
13
+ ## What is Brownfield Development?
14
+
15
+ Brownfield development refers to adding features, fixing bugs, or modernizing existing software projects. Unlike greenfield (new) projects, brownfield work requires understanding existing code, respecting constraints, and ensuring new changes integrate seamlessly without breaking existing functionality.
16
+
17
+ ## When to Use BMAD for Brownfield
18
+
19
+ BMAD-METHOD excels at brownfield development when you need to:
20
+
21
+ - Add significant new features to existing applications
22
+ - Modernize legacy codebases
23
+ - Integrate new technologies or services
24
+ - Refactor complex systems
25
+ - Fix bugs that require architectural understanding
26
+ - Document undocumented systems
27
+
28
+ ## The Complete Brownfield Workflow
29
+
30
+ ### Choose Your Approach
31
+
32
+ #### Approach A: PRD-First (Recommended for Large Codebases/Monorepos)
33
+
34
+ **Best for**: Large codebases, monorepos, or when you know exactly what you want to build
35
+
36
+ 1. **Create PRD First** to define requirements
37
+ 2. **Document only relevant areas** based on PRD needs
38
+ 3. **More efficient** - avoids documenting unused code
39
+
40
+ #### Approach B: Document-First (Good for Smaller Projects)
41
+
42
+ **Best for**: Smaller codebases, unknown systems, or exploratory changes
43
+
44
+ 1. **Document entire system** first
45
+ 2. **Create PRD** with full context
46
+ 3. **More thorough** - captures everything
47
+
48
+ ### Approach A: PRD-First Workflow (Recommended)
49
+
50
+ #### Phase 1: Define Requirements First
51
+
52
+ **In Gemini Web (with your codebase uploaded):**
53
+
54
+ ```bash
55
+ @pm
56
+ *create-doc brownfield-prd
57
+ ```
58
+
59
+ The PM will:
60
+
61
+ - **Ask about your enhancement** requirements
62
+ - **Explore the codebase** to understand current state
63
+ - **Identify affected areas** that need documentation
64
+ - **Create focused PRD** with clear scope
65
+
66
+ **Key Advantage**: The PRD identifies which parts of your monorepo/large codebase actually need documentation!
67
+
68
+ #### Phase 2: Focused Documentation
69
+
70
+ **Still in Gemini Web, now with PRD context:**
71
+
72
+ ```bash
73
+ @analyst
74
+ *document-project
75
+ ```
76
+
77
+ The analyst will:
78
+
79
+ - **Ask about your focus** if no PRD was provided
80
+ - **Offer options**: Create PRD, provide requirements, or describe the enhancement
81
+ - **Reference the PRD/description** to understand scope
82
+ - **Focus on relevant modules** identified in PRD or your description
83
+ - **Skip unrelated areas** to keep docs lean
84
+ - **Generate ONE architecture document** for all environments
85
+
86
+ The analyst creates:
87
+
88
+ - **One comprehensive architecture document** following fullstack-architecture template
89
+ - **Covers all system aspects** in a single file
90
+ - **Easy to copy and save** as `docs/project-architecture.md`
91
+ - **Can be sharded later** in IDE if desired
92
+
93
+ For example, if you say "Add payment processing to user service":
94
+
95
+ - Documents only: user service, API endpoints, database schemas, payment integrations
96
+ - Creates focused source tree showing only payment-related code paths
97
+ - Skips: admin panels, reporting modules, unrelated microservices
98
+
99
+ ### Approach B: Document-First Workflow
100
+
101
+ #### Phase 1: Document the Existing System
102
+
103
+ **Best Approach - Gemini Web with 1M+ Context**:
104
+
105
+ 1. **Go to Gemini Web** (gemini.google.com)
106
+ 2. **Upload your project**:
107
+ - **Option A**: Paste your GitHub repository URL directly
108
+ - **Option B**: Upload up to 1000 files from your src/project folder
109
+ - **Option C**: Zip your project and upload the archive
110
+ 3. **Load the analyst agent**: Upload `dist/agents/analyst.txt`
111
+ 4. **Run documentation**: Type `*document-project`
112
+
113
+ The analyst will generate comprehensive documentation of everything.
114
+
115
+ #### Phase 2: Plan Your Enhancement
116
+
117
+ #### Option A: Full Brownfield Workflow (Recommended for Major Changes)
118
+
119
+ **1. Create Brownfield PRD**
120
+
121
+ ```bash
122
+ @pm
123
+ *create-doc brownfield-prd
124
+ ```
125
+
126
+ The PM agent will:
127
+
128
+ - **Analyze existing documentation** from Phase 1
129
+ - **Request specific enhancement details** from you
130
+ - **Assess complexity** and recommend approach
131
+ - **Create epic/story structure** for the enhancement
132
+ - **Identify risks and integration points**
133
+
134
+ **How PM Agent Gets Project Context**:
135
+
136
+ - In Gemini Web: Already has full project context from Phase 1 documentation
137
+ - In IDE: Will ask "Please provide the path to your existing project documentation"
138
+
139
+ **Key Prompts You'll Encounter**:
140
+
141
+ - "What specific enhancement or feature do you want to add?"
142
+ - "Are there any existing systems or APIs this needs to integrate with?"
143
+ - "What are the critical constraints we must respect?"
144
+ - "What is your timeline and team size?"
145
+
146
+ **2. Create Brownfield Architecture**
147
+
148
+ ```bash
149
+ @architect
150
+ *create-doc brownfield-architecture
151
+ ```
152
+
153
+ The architect will:
154
+
155
+ - **Review the brownfield PRD**
156
+ - **Design integration strategy**
157
+ - **Plan migration approach** if needed
158
+ - **Identify technical risks**
159
+ - **Define compatibility requirements**
160
+
161
+ #### Option B: Quick Enhancement (For Focused Changes)
162
+
163
+ **For Single Epic Without Full PRD**:
164
+
165
+ ```bash
166
+ @pm
167
+ *brownfield-create-epic
168
+ ```
169
+
170
+ Use when:
171
+
172
+ - Enhancement is well-defined and isolated
173
+ - Existing documentation is comprehensive
174
+ - Changes don't impact multiple systems
175
+ - You need quick turnaround
176
+
177
+ **For Single Story**:
178
+
179
+ ```bash
180
+ @pm
181
+ *brownfield-create-story
182
+ ```
183
+
184
+ Use when:
185
+
186
+ - Bug fix or tiny feature
187
+ - Very isolated change
188
+ - No architectural impact
189
+ - Clear implementation path
190
+
191
+ ### Phase 3: Validate Planning Artifacts
192
+
193
+ ```bash
194
+ @po
195
+ *execute-checklist po-master-checklist
196
+ ```
197
+
198
+ The PO ensures:
199
+
200
+ - Compatibility with existing system
201
+ - No breaking changes planned
202
+ - Risk mitigation strategies in place
203
+ - Clear integration approach
204
+
205
+ ### Phase 4: Transition to Development
206
+
207
+ Follow the enhanced IDE Development Workflow:
208
+
209
+ 1. **Ensure documents are in project**:
210
+
211
+ - Copy `docs/prd.md` (or brownfield-prd.md)
212
+ - Copy `docs/architecture.md` (or brownfield-architecture.md)
213
+
214
+ 2. **Shard documents**:
215
+
216
+ ```bash
217
+ @po
218
+ # Ask to shard docs/prd.md
219
+ ```
220
+
221
+ 3. **Development cycle**:
222
+ - **SM** creates stories with integration awareness
223
+ - **Dev** implements with existing code respect
224
+ - **QA** reviews for compatibility and improvements
225
+
226
+ ## Brownfield Best Practices
227
+
228
+ ### 1. Always Document First
229
+
230
+ Even if you think you know the codebase:
231
+
232
+ - Run `document-project` to capture current state
233
+ - AI agents need this context
234
+ - Discovers undocumented patterns
235
+
236
+ ### 2. Respect Existing Patterns
237
+
238
+ The brownfield templates specifically look for:
239
+
240
+ - Current coding conventions
241
+ - Existing architectural patterns
242
+ - Technology constraints
243
+ - Team preferences
244
+
245
+ ### 3. Plan for Gradual Rollout
246
+
247
+ Brownfield changes should:
248
+
249
+ - Support feature flags
250
+ - Plan rollback strategies
251
+ - Include migration scripts
252
+ - Maintain backwards compatibility
253
+
254
+ ### 4. Test Integration Thoroughly
255
+
256
+ Focus testing on:
257
+
258
+ - Integration points
259
+ - Existing functionality (regression)
260
+ - Performance impact
261
+ - Data migrations
262
+
263
+ ### 5. Communicate Changes
264
+
265
+ Document:
266
+
267
+ - What changed and why
268
+ - Migration instructions
269
+ - New patterns introduced
270
+ - Deprecation notices
271
+
272
+ ## Common Brownfield Scenarios
273
+
274
+ ### Scenario 1: Adding a New Feature
275
+
276
+ 1. Document existing system
277
+ 2. Create brownfield PRD focusing on integration
278
+ 3. Architecture emphasizes compatibility
279
+ 4. Stories include integration tasks
280
+
281
+ ### Scenario 2: Modernizing Legacy Code
282
+
283
+ 1. Extensive documentation phase
284
+ 2. PRD includes migration strategy
285
+ 3. Architecture plans gradual transition
286
+ 4. Stories follow strangler fig pattern
287
+
288
+ ### Scenario 3: Bug Fix in Complex System
289
+
290
+ 1. Document relevant subsystems
291
+ 2. Use `brownfield-create-story` for focused fix
292
+ 3. Include regression test requirements
293
+ 4. QA validates no side effects
294
+
295
+ ### Scenario 4: API Integration
296
+
297
+ 1. Document existing API patterns
298
+ 2. PRD defines integration requirements
299
+ 3. Architecture ensures consistent patterns
300
+ 4. Stories include API documentation updates
301
+
302
+ ## Troubleshooting
303
+
304
+ ### "The AI doesn't understand my codebase"
305
+
306
+ **Solution**: Re-run `document-project` with more specific paths to critical files
307
+
308
+ ### "Generated plans don't fit our patterns"
309
+
310
+ **Solution**: Update generated documentation with your specific conventions before planning phase
311
+
312
+ ### "Too much boilerplate for small changes"
313
+
314
+ **Solution**: Use `brownfield-create-story` instead of full workflow
315
+
316
+ ### "Integration points unclear"
317
+
318
+ **Solution**: Provide more context during PRD creation, specifically highlighting integration systems
319
+
320
+ ## Quick Reference
321
+
322
+ ### Brownfield-Specific Commands
323
+
324
+ ```bash
325
+ # Document existing project
326
+ @analyst → *document-project
327
+
328
+ # Create enhancement PRD
329
+ @pm → *create-doc brownfield-prd
330
+
331
+ # Create architecture with integration focus
332
+ @architect → *create-doc brownfield-architecture
333
+
334
+ # Quick epic creation
335
+ @pm → *brownfield-create-epic
336
+
337
+ # Single story creation
338
+ @pm → *brownfield-create-story
339
+ ```
340
+
341
+ ### Decision Tree
342
+
343
+ ```text
344
+ Do you have a large codebase or monorepo?
345
+ ├─ Yes → PRD-First Approach
346
+ │ └─ Create PRD → Document only affected areas
347
+ └─ No → Is the codebase well-known to you?
348
+ ├─ Yes → PRD-First Approach
349
+ └─ No → Document-First Approach
350
+
351
+ Is this a major enhancement affecting multiple systems?
352
+ ├─ Yes → Full Brownfield Workflow
353
+ └─ No → Is this more than a simple bug fix?
354
+ ├─ Yes → brownfield-create-epic
355
+ └─ No → brownfield-create-story
356
+ ```
357
+
358
+ ## Conclusion
359
+
360
+ Brownfield development with BMAD-METHOD provides structure and safety when modifying existing systems. The key is providing comprehensive context through documentation, using specialized templates that consider integration requirements, and following workflows that respect existing constraints while enabling progress.
361
+
362
+ Remember: **Document First, Plan Carefully, Integrate Safely**
@@ -4,7 +4,7 @@ This task guides you through creating a new BMAD agent following the standard te
4
4
 
5
5
  ## Prerequisites
6
6
 
7
- - Agent template: `.bmad-core/templates/agent-tmpl.md`
7
+ - Agent template: `../templates/agent-tmpl.md`
8
8
  - Target directory: `.bmad-core/agents/`
9
9
 
10
10
  ## Steps
@@ -18,7 +18,7 @@ Every expansion pack MUST include a custom BMAD orchestrator agent with sophisti
18
18
 
19
19
  1. **Create Planning Document First**: Before any implementation, create a comprehensive plan for user approval
20
20
  2. **Agent Architecture Standards**: Use YAML-in-Markdown structure with activation instructions, personas, and command systems
21
- 3. **Character Consistency**: Every agent must have a persistent persona with name, communication style, and numbered options protocol
21
+ 3. **Character Consistency**: Every agent must have a persistent persona with name, communication style, and numbered options protocol similar to `expansion-packs/bmad-2d-phaser-game-dev/agents/game-designer.md`
22
22
  4. **Custom Themed Orchestrator**: The orchestrator should embody the domain theme (e.g., Office Manager for medical, Project Lead for tech) for better user experience
23
23
  5. **Core Utilities Required**: ALWAYS include these core files in every expansion pack:
24
24
  - `tasks/create-doc.md` - Document creation from templates
@@ -26,8 +26,8 @@ Every expansion pack MUST include a custom BMAD orchestrator agent with sophisti
26
26
  - `utils/template-format.md` - Template markup conventions
27
27
  - `utils/workflow-management.md` - Workflow orchestration
28
28
  6. **Team and Workflow Requirements**: If pack has >1 agent, MUST include:
29
- - At least one team configuration in `agent-teams/`
30
- - At least one workflow in `workflows/`
29
+ - At least one team configuration in `expansion-packs/{new-expansion}/agent-teams/`
30
+ - At least one workflow in `expansion-packs/{new-expansion}workflows/`
31
31
  7. **Template Sophistication**: Implement LLM instruction embedding with `[[LLM: guidance]]`, conditional content, and variable systems
32
32
  8. **Workflow Orchestration**: Include decision trees, handoff protocols, and validation loops
33
33
  9. **Quality Assurance Integration**: Multi-level checklists with star ratings and ready/not-ready frameworks
@@ -1018,11 +1018,3 @@ Embedded knowledge (automatic):
1018
1018
  - [ ] Template conditional content tested with different scenarios
1019
1019
  - [ ] Workflow decision trees validated with sample use cases
1020
1020
  - [ ] Character interactions tested for consistency and professional authenticity
1021
-
1022
- ```
1023
-
1024
- ```
1025
-
1026
- ```
1027
-
1028
- ```
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "bmad-method",
3
- "version": "4.12.0",
3
+ "version": "4.14.0",
4
4
  "description": "Breakthrough Method of Agile AI-driven Development",
5
5
  "main": "tools/cli.js",
6
6
  "bin": {