Orchestrates BMAD workflows for structured AI-driven development. Use when initializing BMAD in projects, checking workflow status, or routing between 4 phases (Analysis, Planning, Solutioning, Implementation). Manages project configs, tracks progress through project levels 0-4, and coordinates with specialized workflows. Trigger on /workflow-init, /workflow-status, or when users need BMAD setup.
Install
npx skillscat add coco-de/bmad/bmad-master Install via the SkillsCat registry.
BMAD Orchestrator
Purpose: Core orchestrator for the BMAD Method (Breakthrough Method for Agile AI-Driven Development), managing workflows, tracking status, and routing users through structured development phases.
Quick Start
# Initialize BMAD in your project
/workflow-init
# Start a full BMAD workflow
/bmad "Add user authentication"
# Check current progress
/workflow-status
# Run a specific persona review
/bmad:review --persona architect "Review current PR"
# Manually verify a phase gate
/bmad:gate --phase analysisWhen to Use This Skill
Use this skill when:
- User requests
/workflow-initor/init- Initialize BMAD in a project - User requests
/workflow-statusor/status- Check progress and get recommendations - User mentions "BMAD setup" or "start BMAD workflow"
- Project needs structured development methodology
- Coordination between multiple development phases is required
Core Responsibilities
- Project Initialization - Set up BMAD directory structure and configuration
- Status Tracking - Monitor progress across 4 development phases
- Workflow Routing - Direct users to appropriate next steps based on project state
- Progress Management - Maintain workflow status and completion tracking
BMAD Method Overview
4 Development Phases
- Analysis (Optional) - Research, brainstorming, product brief
- Planning (Required) - PRD or Tech Spec based on project complexity
- Solutioning (Conditional) - Architecture design for medium+ projects
- Implementation (Required) - Sprint planning, stories, development
Project Levels
- Level 0: Single atomic change (1 story) - Quick fixes, small tweaks
- Level 1: Small feature (1-10 stories) - Single feature additions
- Level 2: Medium feature set (5-15 stories) - Multiple related features
- Level 3: Complex integration (12-40 stories) - System integrations
- Level 4: Enterprise expansion (40+ stories) - Large-scale projects
Planning Requirements by Level:
- Level 0-1: Tech Spec required, PRD optional/recommended
- Level 2+: PRD required, Tech Spec optional
- Level 2+: Architecture required
Persona Overview
BMAD uses 7 specialized personas across the 4 development phases:
| Persona | Phase | Role |
|---|---|---|
| Analyst | Analysis | Requirements analysis, BDD Gherkin AC authoring |
| Product Manager | Planning | Issue structuring, Story Point estimation |
| Architect | Solutioning | Architecture design review |
| UX Designer | Solutioning | UI/UX design review |
| Frontend Developer | Implementation | Frontend implementation |
| Backend Developer | Implementation | Backend implementation |
| Scrum Master | Implementation | Workflow management |
Workflow Execution Flow
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
â BMAD Workflow â
â âââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ£
â â
â Phase 1: ANALYSIS â
â âââ Analyst review â
â â âââ Requirements analysis â
â â âââ BDD Gherkin AC authoring â
â â âââ Feasibility review â
â âââ Gate: Requirements clarity, scope, AC format â
â â
â Phase 2: PLANNING â
â âââ Product Manager review â
â â âââ Epic/Story structuring â
â â âââ Story Point estimation â
â â âââ Issue creation â
â âââ Gate: Structure, points, labeling, dependencies â
â â
â Phase 3: SOLUTIONING (parallel) â
â âââ Architect review â
â â âââ Clean Architecture â
â â âââ DI structure â
â â âââ API design â
â âââ UX Designer review â
â â âââ Design system compliance â
â â âââ Layout â
â â âââ Interactions â
â âââ Gate: All reviews passed â
â â
â Phase 4: IMPLEMENTATION â
â âââ Frontend Developer â
â âââ Backend Developer â
â âââ Scrum Master (workflow management) â
â âââ Gate: Lint, tests, code review â
â â
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââPhase Gate System
All gates are mandatory - a failed gate blocks progression to the next phase.
Gate FAILED â Feedback provided â Fix issues â Re-review required
Gate PASSED â Automatic progression to next phaseGate Validation Items
| Gate | Required Validations |
|---|---|
| Analysis | Requirements clarity, scope appropriateness, AC in BDD Gherkin format, AC completeness (happy-path + error-handling) |
| Planning | Epic/Story structure, Story Points (1-8), labeling, dependencies |
| Solutioning | Clean Architecture, DI structure, API design, design system compliance, layout, interactions |
| Implementation | Branch naming, lint passing, tests passing, code review completed |
BDD Gherkin Acceptance Criteria
Acceptance criteria must be written in BDD Gherkin Given-When-Then format. The Analysis Gate validates AC format compliance.
Feature: User Authentication
Scenario: Successful login with valid credentials
Given the user is on the login page
When they enter valid credentials
And tap the login button
Then they should be redirected to the home page
Scenario: Login failure with invalid password
Given the user is on the login page
When they enter an invalid password
And tap the login button
Then an error message should be displayedRequirements:
- Every story must have at least one happy-path scenario
- Error-handling scenarios are required for user-facing features
- Each scenario should be testable and unambiguous
Feedback Loop
When a gate fails, the orchestrator provides structured feedback and requires re-review before proceeding.
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
â Phase 3 Gate: FAILED â
â âââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ£
â â
â Rejection reason: â
â - Architect: Service layer bypasses domain abstractions â
â â
â Required actions: â
â 1. Introduce a use case / interactor for the operation â
â 2. Route service calls through the domain layer â
â â
â After fixes, re-review required: â
â /bmad:review --persona architect --retry â
â â
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââRe-review Commands
# Re-review by specific persona
/bmad:review --persona architect --retry
# Re-review entire phase gate
/bmad:gate --phase solutioning --retryEmergency Mode
For production incidents or urgent situations, gates can be streamlined with user approval.
Approval Flow
1. /bmad --emergency "Emergency fix description"
2. Claude requests approval via AskUserQuestion:
âââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
â â ï¸ Emergency Mode Approval Request â
â â
â Emergency mode skips Analysis and Planning gates. â
â Implementation gate (lint, tests) remains mandatory. â
â â
â Do you approve? â
â [Yes, approve] [No, use normal mode] â
âââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
3. If approved: Streamlined workflow proceeds
4. If denied: Falls back to normal BMAD workflow
5. After completion: Post-incident review required within 48 hoursEmergency Mode Constraints
| Item | Emergency Behavior |
|---|---|
| Analysis Gate | Skipped |
| Planning Gate | Skipped |
| Solutioning Gate | Simplified (basic review only) |
| Implementation Gate | Mandatory (lint + tests required) |
| Post-incident Review | Mandatory (within 48 hours, Architect + PM) |
Visual Progress Indicators
Use these ASCII box templates for workflow progress display:
In-Progress Display
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
â BMAD Workflow: "Add user authentication" â
â âââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ£
â â
â Phase 1: ANALYSIS âââââââââââââââââââââââââââââââââ â
â
â Phase 2: PLANNING ââââââââââââââââââââââââââââââââ â
â
â Phase 3: SOLUTIONING âââââââââââââââââââââââââââââ ð â
â Phase 4: IMPLEMENTATION ââââââââââââââââââââââââââ â³ â
â â
â Current: Architect review in progress (UX Designer complete) â
â â
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââCompletion Display
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
â BMAD Workflow Complete: "Add user authentication" â
â âââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ£
â â
â â
Phase 1: ANALYSIS - Approved â
â â
Phase 2: PLANNING - Issue #42 created â
â â
Phase 3: SOLUTIONING - Design approved â
â â
Phase 4: IMPLEMENTATION - PR #45 merged â
â â
â Summary: â
â - Reviews passed: 7/7 personas â
â - Gates passed: 4/4 phases â
â - Retries: 1 (Architect) â
â â
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââAvailable Commands
/workflow-init or /init
Initialize BMAD structure in the current project.
Steps:
Create directory structure:
bmad/ âââ config.yaml âââ agent-overrides/ docs/ âââ bmm-workflow-status.yaml âââ stories/ .claude/commands/bmad/ (if not exists)Collect project information:
- Project name
- Project type (web-app, mobile-app, api, game, library, other)
- Project level (0-4)
Create project config using config.template.yaml
Create workflow status file with conditional requirements based on level:
- Use workflow-status template
- Set PRD: required if level >= 2, else recommended
- Set Tech-spec: required if level <= 1, else optional
- Set Architecture: required if level >= 2, else optional
Display initialization summary and recommend next workflow
Offer to start recommended workflow
Example interaction:
User: /workflow-init
Orchestrator: I'll initialize BMAD for your project.
[Collects info, creates files...]
â BMAD Method initialized!
Project: MyApp
Type: web-app
Level: 2 (Medium feature set)
Configuration: bmad/config.yaml
Status tracking: docs/bmm-workflow-status.yaml
Recommended next step: Create Product Brief with /product-brief
This will help define your product vision and requirements.
Would you like to start with /product-brief?/workflow-status or /status
Check project status and get recommendations for next steps.
Steps:
- Load project config (bmad/config.yaml)
- Load workflow status (docs/bmm-workflow-status.yaml)
- Determine current phase and next recommended workflow
- Display status with visual indicators
- Offer to execute recommended workflow
Status Indicators:
- â = Completed (shows file path)
- â = Required but not started
- â = Current phase
- = Optional/not required
Example interaction:
User: /status
Orchestrator: Checking your project status...
Project: MyApp (Web Application, Level 2)
â Phase 1: Analysis
â product-brief (docs/product-brief-myapp-2025-01-11.md)
â Phase 2: Planning [CURRENT]
â prd (required - NOT STARTED)
- tech-spec (optional)
Phase 3: Solutioning
- architecture (required)
Phase 4: Implementation
- sprint-planning (required)
Recommended next step: Create PRD with /prd command
Would you like to run /prd to create your PRD?If project not initialized:
- Inform user BMAD not detected
- Offer to run
/workflow-init
Workflow Routing Logic
After determining project status, route users to specialized workflows:
- Analysis workflows:
/product-brief,/brainstorm,/research - Planning workflows:
/prd,/tech-spec - UX workflows:
/create-ux-design - Architecture workflows:
/architecture - Sprint workflows:
/sprint-planning,/create-story - Development workflows:
/dev-story,/code-review - Team workflows:
/team-dev,/team-create-stories,/team-review
Recommendation logic:
- If no product-brief and project new â Recommend:
/product-brief - If product-brief complete, no PRD/tech-spec:
- Level 0-1 â Recommend:
/tech-spec - Level 2+ â Recommend:
/prd
- Level 0-1 â Recommend:
- If PRD/tech-spec complete, no architecture, level 2+ â Recommend:
/architecture - If planning complete â Recommend:
/sprint-planning - If sprint active â Recommend:
/create-storyor/dev-story
See REFERENCE.md for detailed routing logic.
Configuration Files
Project Config (bmad/config.yaml)
project_name: "MyApp"
project_type: "web-app" # web-app, mobile-app, api, game, library, other
project_level: 2 # 0-4
output_folder: "docs"
communication_language: "English"
# Optional: docs site sync script (run after BMAD doc generation)
# docs_sync_script: "python3 scripts/sync_project_docs.py"Workflow Status (docs/bmm-workflow-status.yaml)
Tracks completion of each workflow with status values:
"optional"- Can be skipped"recommended"- Strongly suggested"required"- Must be completed"{file-path}"- Completed (shows output file)"skipped"- Explicitly skipped
See templates/config.template.yaml for full template.
Project Docs Integration
BMAD-generated documents are stored in the docs/ directory with a standardized naming pattern:
docs/{type}-{feature-slug}-{YYYY-MM-DD}.md
Examples:
docs/prd-user-auth-2025-03-15.md
docs/architecture-user-auth-2025-03-15.md
docs/tech-spec-user-auth-2025-03-15.md| Document Type | File Prefix |
|---|---|
| Product Brief | product-brief- |
| PRD | prd- |
| Tech Spec | tech-spec- |
| UX Design | ux-design- |
| Architecture | architecture- |
Docs Site Sync (Optional)
If your project has a docs site, configure automatic sync in bmad/config.yaml:
# bmad/config.yaml
docs_sync_script: "python3 scripts/sync_project_docs.py"When configured, the orchestrator will offer to run the sync script after generating BMAD documents. The sync script is project-specific and should handle copying docs to the appropriate location for your docs site.
Helper Scripts
Execute via Bash tool:
init-project.sh - Automated project initialization
bash scripts/init-project.sh --name "MyApp" --type web-app --level 2check-status.sh - Display current workflow status
bash scripts/check-status.shvalidate-config.sh - Validate YAML configuration
bash scripts/validate-config.sh bmad/config.yaml
See scripts documentation for details.
Error Handling
Config missing:
- Suggest
/workflow-init - Explain BMAD not initialized
Invalid YAML:
- Show error location
- Offer to fix or reinitialize
Template missing:
- Use inline fallback
- Log warning
- Continue operation
Status file inconsistent:
- Validate against project level
- Offer to regenerate
Integration with Other Skills
This orchestrator coordinates with specialized BMAD skills:
business-analyst- Analysis phase workflowsproduct-manager- Planning phase workflowssystem-architect- Architecture designscrum-master- Sprint and story managementdeveloper- Development workflows
When routing to these skills, pass context:
- Current project config
- Workflow status
- Project level
- Output folder location
Token Optimization
- Use script automation for repetitive tasks
- Reference REFERENCE.md for detailed logic
- Load files only when needed
- Keep status displays concise
- Delegate detailed work to specialized skills
Subagent Strategy
This skill leverages parallel subagents to maximize context utilization (each agent has up to 1M tokens on Claude Sonnet 4.6 / Opus 4.6).
Subagent Types
| Type | Model | Tools | Use Case |
|---|---|---|---|
general-purpose |
Inherited | All tools | Full-capability subagent (default) |
Explore |
Haiku (fast) | Read-only | Codebase exploration, search |
Plan |
Inherited | Read-only | Architecture design, planning |
Token Budget Guidelines
| Category | Budget | Use Case |
|---|---|---|
| Research/Exploration | ~500K | Codebase analysis, pattern investigation |
| Generation/Writing | ~300K | Code generation, document writing |
| Validation/Testing | ~80K | Lint, test execution |
Parallel Execution: Solutioning Phase
Architect and UX Designer review simultaneously:
Agent({
subagent_type: "general-purpose",
description: "Architect review",
prompt: "Review clean architecture, DI structure, API design...",
})
// simultaneously
Agent({
subagent_type: "general-purpose",
description: "UX Designer review",
prompt: "Review design system compliance, layout, interactions...",
})Worktree Isolation: Implementation Phase
When backend and frontend work is independent, use worktree isolation to prevent file conflicts:
Agent({
subagent_type: "general-purpose",
description: "Backend implementation",
prompt: "Implement API endpoints...",
isolation: "worktree", // Independent git worktree
})
// simultaneously
Agent({
subagent_type: "general-purpose",
description: "Frontend implementation",
prompt: "Implement UI with mock data...",
isolation: "worktree", // Prevents file conflicts
})
isolation: "worktree": Creates a temporary git worktree for isolated work.
If no changes are made, the worktree is auto-cleaned. Otherwise, the worktree path and branch are returned.
Background Execution
For long-running research or analysis tasks:
Agent({
subagent_type: "Explore",
description: "Codebase exploration",
prompt: "Analyze existing patterns...",
run_in_background: true, // Returns output_file path
})
// Check progress (non-blocking)
TaskOutput(task_id: "...", block: false)Workflow Status Check Workflow
Pattern: Fan-Out Research
Agents: 3-4 parallel agents
| Agent | Task | Output |
|---|---|---|
| Agent 1 | Check project config and validate structure | bmad/outputs/config-status.md |
| Agent 2 | Analyze workflow status file and phase completion | bmad/outputs/workflow-status.md |
| Agent 3 | Scan docs directory for completed artifacts | bmad/outputs/artifacts-status.md |
| Agent 4 | Generate recommendations based on project level | bmad/outputs/recommendations.md |
Coordination:
- Launch all agents with shared project context
- Each agent writes status findings to designated output
- Main context synthesizes results into unified status report
- Display visual status indicators and next steps
Project Initialization Workflow
Pattern: Parallel Section Generation
Agents: 3 parallel agents
| Agent | Task | Output |
|---|---|---|
| Agent 1 | Create directory structure and validate paths | bmad/outputs/directory-setup.md |
| Agent 2 | Generate project config from template | bmad/config.yaml |
| Agent 3 | Generate workflow status file with level-based requirements | docs/bmm-workflow-status.yaml |
Coordination:
- Gather project information from user (sequential)
- Launch parallel agents to create structures and configs
- Main context validates all outputs and displays summary
Example Subagent Prompt
Task: Analyze workflow status and determine current phase
Context: Read bmad/config.yaml and docs/bmm-workflow-status.yaml
Objective: Identify completed workflows, current phase, and required next steps
Output: Write analysis to bmad/outputs/workflow-status.md
Deliverables:
1. List of completed workflows with file paths
2. Current phase determination
3. Required vs optional next workflows
4. Blocking issues or missing dependencies
Constraints:
- Use project level to determine requirements
- Flag any inconsistencies in status fileRelated Commands
| Command | Description |
|---|---|
/bmad |
Full BMAD workflow |
/bmad:review |
Single persona review |
/bmad:team-review |
Parallel multi-persona review (Agent Teams) |
/bmad:status |
Check workflow status |
/bmad:gate |
Manual gate verification |
/workflow-init |
Initialize BMAD in project |
/workflow-status |
Check project progress |
Notes for Claude
- This is the entry point for BMAD workflows
- Always check if project is initialized before operations
- Maintain phase-based progression (don't skip required phases)
- Use TaskCreate for multi-step initialization tracking
- Keep responses focused and actionable
- Hand off to specialized skills for detailed workflows
- Update workflow status after completing workflows
- Display visual progress indicators at workflow milestones
- Enforce phase gates - never allow progression past a failed gate
- When emergency mode is requested, always get explicit user approval first
Quick Reference
- Detailed routing logic: REFERENCE.md
- Workflow phases: resources/workflow-phases.md
- Config template: templates/config.template.yaml
- Init script: scripts/init-project.sh
- Status script: scripts/check-status.sh