ldy26098
/
BMAD-METHOD
mirror of https://github.com/bmad-code-org/BMAD-METHOD.git
1
0
Fork 0
Code Packages Projects Releases Wiki Activity

feat: Enhance agent capabilities with new commands and update documentation for BMAD standards

This commit is contained in:
Peet Stander 2025-06-05 13:48:38 +02:00
parent 359e3b8b7f
commit 035a936f2e
11 changed files with 289 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
.augment-guidelines
View File

@ -59,6 +59,11 @@ Immediate Action Commands:
- `*exit`: Immediately abandon the current agent or party-mode and drop to base BMad Orchestrator
- `*tasks`: List the tasks available to the current agent, along with a description.
- `*party`: This enters group chat with all available agents. You will roleplay all agent personas as necessary
- `*mpcs`: List available Machine-Powered Capabilities for the current agent
- `*perplexity [query]`: Execute a web search using Perplexity (if available for current agent)
- `*github [query]`: Search code repositories on GitHub (if available for current agent)
- `*firecrawl [query]`: Perform advanced data analysis with Firecrawl (if available for current agent)
- `*dalle [prompt]`: Generate an image using DALL-E (if available for current agent)
## Global Output Requirements Apply to All Personas

5
.github/copilot-instructions.md vendored
View File

@ -59,6 +59,11 @@ Immediate Action Commands:
- `*exit`: Immediately abandon the current agent or party-mode and drop to base BMad Orchestrator
- `*tasks`: List the tasks available to the current agent, along with a description.
- `*party`: This enters group chat with all available agents. You will roleplay all agent personas as necessary
- `*mpcs`: List available Machine-Powered Capabilities for the current agent
- `*perplexity [query]`: Execute a web search using Perplexity (if available for current agent)
- `*github [query]`: Search code repositories on GitHub (if available for current agent)
- `*firecrawl [query]`: Perform advanced data analysis with Firecrawl (if available for current agent)
- `*dalle [prompt]`: Generate an image using DALL-E (if available for current agent)
## Global Output Requirements Apply to All Personas

3
.gitignore vendored
View File

@ -18,3 +18,6 @@ build/
# VSCode settings
.vscode/
.github/
CLAUDE.md

253
bmad-agent/commands.md
View File

@ -84,11 +84,100 @@
4. `*Dev` - Implement the module stories
5. `*QA create-test-plan` - Create test plan for the new module
## File Organization Guidelines
### Directory Structure Overview
The BMAD Method uses a **hybrid file organization system** that evolved from earlier versions:
| Directory | Purpose | Contents | Auto-Generated |
|-----------|---------|----------|----------------|
| `docs/` | **Human Deliverables** | Project Briefs, PRDs, Architecture docs, Stories, API docs | ❌ Manual |
| `.ai/` | **Agent Knowledge & Logs** | Project context, tech stack, data models, agent logs | ✅ Auto-created |
| `bmad-agent/` | **BMAD System Files** | Personas, tasks, templates, configuration | ❌ Manual |
### What Goes Where
**`docs/` Directory - Project Documentation:**
- Project Briefs (from Analyst)
- Product Requirements Documents (from PM)
- Architecture Documents (from Architect)
- UX/UI Specifications (from Design Architect)
- User Stories (in `docs/stories/`)
- API Reference documentation
- Sharded documents (broken down from large docs)
### Standard File Naming Convention
**All BMAD Method core documents MUST use lowercase filenames with hyphens:**
| Document Type | Standard Filename | Agent Responsible |
|---------------|-------------------|-------------------|
| Project Brief | `project-brief.md` | Analyst |
| Product Requirements | `prd.md` | PM |
| Architecture | `architecture.md` | Architect |
| Frontend Architecture | `frontend-architecture.md` | Design Architect |
| UX/UI Specification | `uxui-spec.md` | Design Architect |
| Technology Stack | `tech-stack.md` | Architect |
| Data Models | `data-models.md` | Architect/Data Scientist |
| API Reference | `api-reference.md` | Architect |
| Deployment Guide | `deployment-guide.md` | DevOps |
| Test Plan | `test-plan.md` | QA |
| User Stories | `{epic-num}.{story-num}.story.md` | SM |
| Epic Files | `epic-{id}.md` | SM (from sharding) |
### Date Generation Standards
**All agents MUST use actual current dates, not placeholders:**
- Use format: `YYYY-MM-DD` for dates (e.g., `2024-01-15`)
- Use format: `YYYY-MM-DD HH:MM` for timestamps (e.g., `2024-01-15 14:30`)
- **NEVER use placeholders** like `{DATE}`, `[DATE]`, or `TBD`
- **ALWAYS generate actual current date** when creating documents
**`.ai/` Directory - Agent Intelligence:**
- `project-context.md` - Project goals, terminology, domain knowledge
- `tech-stack.md` - Technologies, frameworks, patterns in use
- `data-models.md` - Data structures and analytics approaches
- `deployment-info.md` - Infrastructure and deployment details
- `knowledge-versions.md` - Version history of knowledge updates
- Agent working files (test-issues.md, deployment-history.md, etc.)
### System Evolution Note
**If your project only has `docs/`:** You're likely using an older configuration or haven't run the knowledge update task yet. The current system automatically creates and manages the `.ai/` directory when you run `*BMAD Update Agent Knowledge`.
### Updating Agent Knowledge
#### When to Run Knowledge Updates
**Required Updates (Major Phases):**
- After Project Brief completion
- After PRD creation or major updates
- After Architecture design or significant changes
- After adding new modules or major features
- After tech stack changes
**Frequency Guidelines:**
- **New Projects**: 3-4 times during initial setup
- **Active Development**: Every 2-4 weeks or when major changes occur
- **Mature Projects**: Monthly or when adding significant features
**DON'T Update After:**
- Individual story completions
- Small bug fixes
- Routine development tasks
- Individual test runs
#### Running the Update
1. Complete a project phase (Brief, PRD, Architecture)
2. `*BMAD Update Agent Knowledge` - Extract and distribute knowledge
3. Verify knowledge files in `.ai/` directory
3. **Automatic Results:**
- Creates `.ai/` directory if it doesn't exist
- Generates knowledge files from your `docs/` content
- Updates agent customization strings
- Uses semantic versioning (Major.Minor.Patch) to track changes
- Agents automatically read from `.ai/` for project context
### Creating Custom Agents
@ -106,6 +195,168 @@
- [Task Name](task-file.md)
```
## Common Scenarios
### 1. Complete Project Initialization Flow
**Purpose:** Start a new project from scratch through to development
| Step | Agent | Command | Description |
|------|-------|---------|-------------|
| 1 | Analyst | `*Analyst Create Project Brief` | Brainstorm and research project concept |
| 2 | PM | `*PM Create PRD` | Create Product Requirements Document with epics and stories |
| 3 | Architect | `*Architect Create Architecture` | Design system architecture based on PRD |
| 4 | Design Architect | `*Design Architect Create Frontend Architecture` | Design UI/UX architecture (if applicable) |
| 5 | Design Architect | `*Design Architect Create UXUI Spec` | Create detailed UI/UX specifications |
| 6 | BMAD | `*BMAD Update Agent Knowledge` | Update all agents with project knowledge |
| 7 | PO | `*PO organize` | Organize and validate all documentation |
| 8 | SM | `*SM doc-shard` | Break down large documents into manageable pieces |
| 9 | SM | `*SM create` | Create first implementation story |
| 10 | Dev | `*Dev` | Implement the story |
| 11 | QA | `*QA create-test-plan` | Create test plan for the story |
| 12 | QA | `*QA run-tests` | Execute tests for the implementation |
| 13 | DevOps | `*DevOps deploy` | Deploy the implementation |
**Special Considerations:**
- Run `*BMAD Update Agent Knowledge` after each major phase
- Consider using `*perplexity` during research phases
- For UI-heavy projects, add `*dalle` for mockup generation after step 5
### 2. Brownfield Project Takeover
**Purpose:** Integrate BMAD Method into an existing project
| Step | Agent | Command | Description |
|------|-------|---------|-------------|
| 1 | Analyst | `*Analyst Analyze Existing Project` | Document current state and challenges |
| 2 | PM | `*PM Reverse Engineer PRD` | Create PRD based on existing functionality |
| 3 | Architect | `*Architect Document Current Architecture` | Map out existing architecture |
| 4 | SM | `*SM doc-shard` | Break down documentation into manageable pieces |
| 5 | BMAD | `*BMAD Update Agent Knowledge` | Update all agents with project knowledge |
| 6 | PO | `*PO audit` | Identify documentation gaps |
| 7 | SM | `*SM create` | Create first enhancement story |
| 8 | Dev | `*Dev` | Implement the enhancement |
**Special Considerations:**
- Use `*github` to search for patterns in the existing codebase
- Consider `*SM pivot` if significant course correction is needed
- Create a project structure document if one doesn't exist
### 3. Adding New Module to Existing Project
**Purpose:** Extend a project beyond original PRD scope
| Step | Agent | Command | Description |
|------|-------|---------|-------------|
| 1 | Analyst | `*Analyst Research Module Requirements` | Research requirements for new module |
| 2 | PM | `*PM update-prd {module-name}` | Update PRD with new module requirements |
| 3 | Architect | `*Architect module-design {module-name}` | Design new module architecture |
| 4 | Design Architect | `*Design Architect Update Frontend Architecture` | Update UI/UX for new module (if applicable) |
| 5 | BMAD | `*BMAD Update Agent Knowledge` | Update all agents with new module knowledge |
| 6 | SM | `*SM doc-shard` | Break down module documentation |
| 7 | SM | `*SM create` | Create first module implementation story |
| 8 | Dev | `*Dev` | Implement the module story |
| 9 | QA | `*QA create-test-plan` | Create test plan for the module |
| 10 | DevOps | `*DevOps infra-plan` | Plan infrastructure changes for new module |
**Special Considerations:**
- Ensure integration points with existing modules are clearly defined
- Consider impact on existing architecture and data models
- Update knowledge files to include new module terminology
### 4. UI Redesign Workflow
**Purpose:** Implement frontend changes with minimal backend modifications
| Step | Agent | Command | Description |
|------|-------|---------|-------------|
| 1 | Design Architect | `*Design Architect Analyze Current UI` | Document current UI state and issues |
| 2 | PM | `*PM Create UI PRD` | Create UI-focused requirements document |
| 3 | Design Architect | `*Design Architect Create UXUI Spec` | Create detailed UI/UX specifications |
| 4 | Design Architect | `*Design Architect Create Frontend Architecture` | Update frontend architecture |
| 5 | Design Architect | `*Design Architect Create AI Frontend Prompt` | Create prompt for AI UI generation |
| 6 | BMAD | `*BMAD Update Agent Knowledge` | Update agents with UI knowledge |
| 7 | SM | `*SM create` | Create UI implementation story |
| 8 | Dev | `*Dev` | Implement UI changes |
| 9 | QA | `*QA create-test-plan` | Create UI-focused test plan |
**Special Considerations:**
- Use `*dalle` for UI mockup generation
- Focus on component-based architecture for reusability
- Consider accessibility requirements in specifications
### 5. API Integration Project
**Purpose:** Integrate external APIs into an existing project
| Step | Agent | Command | Description |
|------|-------|---------|-------------|
| 1 | Analyst | `*Analyst Research API` | Research API capabilities and limitations |
| 2 | PM | `*PM Create API Integration PRD` | Document API integration requirements |
| 3 | Architect | `*Architect Design API Integration` | Design integration architecture |
| 4 | BMAD | `*BMAD Update Agent Knowledge` | Update agents with API knowledge |
| 5 | SM | `*SM create` | Create API integration story |
| 6 | Dev | `*Dev` | Implement API integration |
| 7 | QA | `*QA create-test-plan` | Create API testing plan |
| 8 | DevOps | `*DevOps update-env` | Update environment with API credentials |
**Special Considerations:**
- Use `*perplexity` to research API best practices
- Create mock API responses for testing
- Document rate limits and fallback strategies
### 6. Database Migration Project
**Purpose:** Migrate from one database technology to another
| Step | Agent | Command | Description |
|------|-------|---------|-------------|
| 1 | Analyst | `*Analyst Research Database Options` | Research database technologies |
| 2 | Architect | `*Architect Design Database Migration` | Design migration architecture |
| 3 | Data Scientist | `*Data Scientist analyze` | Analyze data patterns and migration challenges |
| 4 | PM | `*PM Create Migration PRD` | Document migration requirements and phases |
| 5 | BMAD | `*BMAD Update Agent Knowledge` | Update agents with database knowledge |
| 6 | SM | `*SM create` | Create database migration story |
| 7 | Dev | `*Dev` | Implement migration code |
| 8 | QA | `*QA create-test-plan` | Create data validation test plan |
| 9 | DevOps | `*DevOps infra-plan` | Plan infrastructure changes for new database |
**Special Considerations:**
- Create data validation strategies for before and after migration
- Plan for rollback scenarios
- Consider performance testing with representative data volumes
### 7. Performance Optimization Project
**Purpose:** Improve performance of an existing application
| Step | Agent | Command | Description |
|------|-------|---------|-------------|
| 1 | Analyst | `*Analyst Performance Analysis` | Identify performance bottlenecks |
| 2 | Architect | `*Architect Performance Optimization Plan` | Design optimization strategy |
| 3 | PM | `*PM Create Optimization PRD` | Document optimization requirements |
| 4 | BMAD | `*BMAD Update Agent Knowledge` | Update agents with performance knowledge |
| 5 | SM | `*SM create` | Create optimization story |
| 6 | Dev | `*Dev` | Implement optimizations |
| 7 | QA | `*QA create-performance-test-plan` | Create performance test plan |
| 8 | Data Scientist | `*Data Scientist analyze-metrics` | Analyze performance metrics |
**Special Considerations:**
- Establish performance baselines before changes
- Use `*github` to research optimization patterns
- Consider both frontend and backend optimizations
### Best Practices for Scenario Execution
**Documentation First:** Complete documentation phases before implementation
**Knowledge Updates:** Run `*BMAD Update Agent Knowledge` after each major phase
**Incremental Implementation:** Create and implement stories one at a time
**Regular Testing:** Integrate QA testing throughout the process
**Feedback Loops:** Use `*SM pivot` if significant course correction is needed
**MPC Integration:** Leverage appropriate MPCs for each scenario:
- Research: `*perplexity`
- Code patterns: `*github`
- Data analysis: `*firecrawl`
- UI visualization: `*dalle`
## Documentation Management
| Command | Description |

4
bmad-agent/personas/dev.ide.md
View File

@ -17,7 +17,7 @@
MUST review and use:
- `Assigned Story File`: `docs/stories/{epicNumber}.{storyNumber}.story.md`
- `Assigned Story File`: `docs/stories/{epic-num}.{story-num}.story.md` (BMAD standard naming)
- `Project Structure`: `docs/project-structure.md`
- `Operational Guidelines`: `docs/operational-guidelines.md` (Covers Coding Standards, Testing Strategy, Error Handling, Security)
- `Technology Stack`: `docs/tech-stack.md`
@ -47,7 +47,7 @@ MUST review and use:
a. HALT feature implementation concerning the dependency.
b. In story file: document need & strong justification (benefits, alternatives).
c. Ask user for explicit approval for this dependency.
d. ONLY upon user's explicit approval (e.g., "User approved X on YYYY-MM-DD"), document it in the story file and proceed.
d. ONLY upon user's explicit approval (e.g., "User approved X on [current date in YYYY-MM-DD format]"), document it in the story file and proceed.
- **Debugging Protocol:**
- For temporary debug code (e.g., extensive logging):
a. MUST log in `Debugging Log` _before_ applying: include file path, change description, rationale, expected outcome. Mark as 'Temp Debug for Story X.Y'.

11
bmad-agent/tasks/agent-knowledge-update-task.md
View File

@ -4,11 +4,11 @@
Extract key project information from existing documentation and update agent knowledge files and customization strings to make all BMAD agents more effective with project-specific context.
## Input Required
- Project Brief (if available)
- PRD (if available)
- Architecture document (if available)
- Tech Stack document (if available)
- Other project artifacts
- Project Brief (`docs/project-brief.md` if available)
- PRD (`docs/prd.md` if available)
- Architecture document (`docs/architecture.md` if available)
- Tech Stack document (`docs/tech-stack.md` if available)
- Other project artifacts following BMAD naming standards
## Steps
@ -54,6 +54,7 @@ Extract key project information from existing documentation and update agent kno
5. **Update Knowledge Version History**
- Create or update `.ai/knowledge-versions.md` using the template at `templates/knowledge-version-tmpl.md`
- **CRITICAL: Use actual current date in YYYY-MM-DD format, never leave date placeholders**
- Increment version number appropriately based on change significance:
- Major: Significant changes to project direction or core architecture
- Minor: New features or components added to existing architecture

4
bmad-agent/tasks/create-knowledge-files.js
View File

@ -306,7 +306,9 @@ function main() {
## Additional Resources
- Project documentation is located in the \`docs/\` directory
- Architecture diagrams can be found in \`docs/architecture.md\`
- Architecture diagrams can be found in \`docs/architecture.md\` (BMAD standard naming)
- Project brief is in \`docs/project-brief.md\`
- PRD is in \`docs/prd.md\`
`;
updateKnowledgeFile(PROJECT_CONTEXT_PATH, projectContextContent);

2
bmad-agent/tasks/create-next-story-task.md
View File

@ -82,7 +82,7 @@ To identify the next logical story based on project progress and epic definition
### 5. Populate Story Template with Full Context
- Create a new story file: `docs/stories/{epicNum}.{storyNum}.story.md`.
- Create a new story file: `docs/stories/{epic-num}.{story-num}.story.md` (using BMAD standard naming with hyphens).
- Use the Story Template to structure the file.
- Fill in:
- Story `{EpicNum}.{StoryNum}: {Short Title Copied from Epic File}`

6
bmad-agent/templates/doc-sharding-tmpl.md
View File

@ -6,7 +6,7 @@ This plan directs the agent on how to break down large source documents into sma
## 1. Source Document: PRD (Project Requirements Document)
- **Note to Agent:** Confirm the exact filename of the PRD with the user (e.g., `PRD.md`, `ProjectRequirements.md`, `prdx.y.z.md`).
- **Note to Agent:** The PRD should be named `prd.md` per BMAD naming standards. If using a different name, confirm with user and suggest renaming to standard format.
### 1.1. Epic Granulation
@ -19,7 +19,7 @@ This plan directs the agent on how to break down large source documents into sma
## 2. Source Document: Main Architecture Document
- **Note to Agent:** Confirm the exact filename with the user (e.g., `architecture.md`, `SystemArchitecture.md`).
- **Note to Agent:** The Architecture document should be named `architecture.md` per BMAD naming standards. If using a different name, confirm with user and suggest renaming to standard format.
### 2.1. Core Architecture Granules
@ -65,7 +65,7 @@ This plan directs the agent on how to break down large source documents into sma
## 3. Source Document(s): Front-End Specific Documentation
- **Note to Agent:** Confirm filenames with the user (e.g., `front-end-architecture.md`, `front-end-spec.md`, `ui-guidelines.md`). Multiple FE documents might exist.
- **Note to Agent:** Frontend documents should follow BMAD naming standards: `frontend-architecture.md`, `uxui-spec.md`. If using different names, confirm with user and suggest renaming to standard format.
### 3.1. Front-End Granules

10
bmad-agent/templates/knowledge-version-tmpl.md
View File

@ -1,10 +1,12 @@
# Knowledge Version History
## Current Version: {VERSION_NUMBER}
**Last Updated:** {DATE}
**Last Updated:** [AGENT: Use current date in YYYY-MM-DD format]
**Updated By:** {AGENT_NAME}
**Change Type:** {MAJOR|MINOR|PATCH}
**IMPORTANT:** When using this template, replace {DATE} placeholders with actual current dates in YYYY-MM-DD format. Never leave date placeholders in final documents.
## Change Summary
{BRIEF_DESCRIPTION_OF_CHANGES}
@ -40,6 +42,6 @@
| Version | Date | Updated By | Change Type | Summary |
|---------|------|------------|-------------|---------|
| {VERSION_NUMBER} | {DATE} | {AGENT_NAME} | {CHANGE_TYPE} | {BRIEF_SUMMARY} |
| {PREVIOUS_VERSION} | {DATE} | {AGENT_NAME} | {CHANGE_TYPE} | {BRIEF_SUMMARY} |
| {INITIAL_VERSION} | {DATE} | {AGENT_NAME} | Initial | Initial knowledge base creation |
| {VERSION_NUMBER} | [AGENT: Current date YYYY-MM-DD] | {AGENT_NAME} | {CHANGE_TYPE} | {BRIEF_SUMMARY} |
| {PREVIOUS_VERSION} | [AGENT: Previous date YYYY-MM-DD] | {AGENT_NAME} | {CHANGE_TYPE} | {BRIEF_SUMMARY} |
| {INITIAL_VERSION} | [AGENT: Initial date YYYY-MM-DD] | {AGENT_NAME} | Initial | Initial knowledge base creation |

4
bmad-agent/templates/story-tmpl.md
View File

@ -1,4 +1,6 @@
# Story {EpicNum}.{StoryNum}: {Short Title Copied from Epic File}
# Story {epic-num}.{story-num}: {Short Title Copied from Epic File}
**IMPORTANT:** Use BMAD standard naming: {epic-num}.{story-num} with hyphens, not {EpicNum}.{StoryNum}
## Status: { Draft | Approved | InProgress | Review | Done }