BMAD-METHOD/bmad/bmm/workflows/workflow-status/init/instructions.md

Workflow Init - Project Setup Instructions

The workflow execution engine is governed by: {project-root}/bmad/core/tasks/workflow.xml You MUST have already loaded and processed: workflow-init/workflow.yaml Communicate in {communication_language} with {user_name}

Welcome to BMad Method, {user_name}!

Quick scan for context (do NOT analyze in depth yet):

• Check for codebase: src/, lib/, package.json, .git, etc.
• Check for BMM artifacts: PRD, epics, stories, tech-spec, architecture docs
• Store what was found but do NOT infer project details yet

What's your project called? {{#if project_name}}(Config shows: {{project_name}}){{/if}} Set project_name project_name

I found some existing work here. Let me understand what you're working on: **Planning Documents Found:** {{#each artifacts}} - {{artifact_name}} ({{artifact_type}}, {{story_count}} stories, modified {{date}}) {{/each}} **Codebase Found:** - Source code in: {{source_dirs}} - Tech stack: {{detected_tech_stack}} {{#if git_history}} - Git history: {{commit_count}} commits, last commit {{last_commit_date}} {{/if}}

Looking at what I found, are these:

a) Works in progress you're finishing - continuing the work described in these documents b) Documents from a previous effort - you're starting something NEW and different now c) The proposed work you're about to start - these describe what you want to do d) None of these - let me explain what I'm actually working on

Your choice [a/b/c/d]:

User is continuing old work - analyze artifacts to get details Set continuing_old_work = true Go to Step 2 (Analyze artifacts for details) User is doing NEW work - old artifacts are just context Set continuing_old_work = false Go to Step 3 (Ask about NEW work) Artifacts describe proposed work Set continuing_old_work = true Go to Step 2 (Analyze artifacts for details) User will explain their situation Go to Step 3 (Ask about their work) I don't see any existing code or planning documents. Looks like we're starting fresh! Go to Step 3 (Ask about their work) Analyze found artifacts in detail: Extract project type from content (game vs software) Count stories/epics to estimate level: - Level 0: 1 story - Level 1: 1-10 stories - Level 2: 5-15 stories - Level 3: 12-40 stories - Level 4: 40+ stories Detect field type from codebase presence (greenfield vs brownfield)

Based on the artifacts you're continuing, I'm suggesting Level {{project_level}} because I found {{story_count}} stories across {{epic_count}} epics.

Here's the complexity scale for reference:

{{field_type}} Project Levels:

• Level 0 - Single atomic change (1 story) - bug fixes, typos, minor updates
• Level 1 - Small feature (1-10 stories) - simple additions, isolated features
• Level 2 - Medium feature set (5-15 stories) - dashboards, multiple related features
• Level 3 - Complex integration (12-40 stories) - platform features, major integrations
• Level 4 - Enterprise expansion (40+ stories) - multi-tenant, ecosystem changes

My suggestion: Level {{project_level}} {{field_type}} {{project_type}} project

Does this match what you're working on? (y/n or tell me what's different)

Use analyzed values Go to Step 4 (Load workflow path) Update values based on user corrections Updated to: Level {{project_level}} {{field_type}} {{project_type}}. Correct? (y/n) Go to Step 4 (Load workflow path)

project_name project_type project_level field_type

Tell me about what you're working on. What's the goal?

Analyze user's description using keyword detection:

• Level 0 keywords: "fix", "bug", "typo", "small change", "update", "patch", "one file"
• Level 1 keywords: "simple", "basic", "small feature", "add", "minor", "single feature"
• Level 2 keywords: "dashboard", "several features", "admin panel", "medium", "feature set"
• Level 3 keywords: "platform", "integration", "complex", "system", "architecture"
• Level 4 keywords: "enterprise", "multi-tenant", "multiple products", "ecosystem", "phased"

Make initial determination:

• project_type (game or software)
• project_level (0-4) - tentative based on keywords
• field_type (greenfield or brownfield)
• confidence (high/medium/low) - based on clarity of description

Thanks! Let me ask a few clarifying questions to make sure I route you correctly:

1. Roughly how many distinct features or changes do you think this involves?

• Just one thing (e.g., fix a bug, add one button, update one API)
• A small feature (2-5 related changes)
• Several features (5-15 related things)
• A major addition (15-40 things to do)
• A large initiative (40+ changes across many areas)

Adjust project_level based on response

2. How much of the existing codebase will this touch?

• Single file or small area
• One module or component
• Multiple modules (2-4 areas)
• Many modules with integration needs
• System-wide changes

Validate and adjust project_level based on scope

3. Is this a game or a software application? Set project_type based on response I see you have existing code here. Are you:

1. Adding to or modifying the existing codebase (brownfield)
2. Starting fresh - the existing code is just a scaffold/template (greenfield)
3. Something else - let me clarify

Your choice [1/2/3]:

Set field_type = "brownfield" Set field_type = "greenfield" Got it - treating as greenfield despite the scaffold. Please explain your situation: Analyze explanation and set field_type accordingly

Build reasoning for suggestion Store detected_indicators (keywords, scope indicators, complexity signals)

Based on what you've described, I'm suggesting Level {{project_level}} because:

{{reasoning}} (detected: {{detected_indicators}})

Here's the complexity scale for reference:

{{field_type}} Project Levels:

• Level 0 - Single atomic change (1 story) - bug fixes, typos, minor updates, single file changes
• Level 1 - Small feature (1-10 stories) - simple additions, isolated features, one module
• Level 2 - Medium feature set (5-15 stories) - dashboards, multiple related features, several modules
• Level 3 - Complex integration (12-40 stories) - platform features, major integrations, architectural changes
• Level 4 - Enterprise expansion (40+ stories) - multi-tenant, ecosystem changes, system-wide initiatives

My suggestion: Level {{project_level}} {{field_type}} {{project_type}} project

Does this match what you're working on? (y/n or tell me what's different)

Use determined values Go to Step 4 (Load workflow path) Update values based on corrections Updated to: Level {{project_level}} {{field_type}} {{project_type}} Does that look right now? (y/n) If yes, go to Step 4. If no, ask what needs adjustment and repeat.

project_name project_type project_level field_type

Determine path file based on selections: Load {path_files}/game-design.yaml Set workflow_path_file = "game-design.yaml" Build filename: {field_type}-level-{project_level}.yaml Load {path_files}/{field_type}-level-{project_level}.yaml Set workflow_path_file = constructed filename

Parse workflow path file to extract phases and workflows workflow_path_file

Parse the loaded workflow path file and extract all workflows

For each phase in the path file:

• Extract phase number and name
• Extract all workflows in that phase
• For each workflow, determine its status type:
  • required: true → status = "required"
  • recommended: true → status = "recommended"
  • conditional: "if_has_ui" → status = "conditional"
  • optional: true → status = "optional"
  • Default if not specified → status = "required"

Build the workflow_items list in this format:

For each phase:

1. Add comment header: # Phase {n}: {Phase Name}
2. For each workflow in phase:
   • Add entry: {workflow-id}: {status}
3. Add blank line between phases

Example structure:

  # Phase 1: Analysis
  brainstorm-project: optional
  research: optional
  product-brief: recommended

  # Phase 2: Planning
  prd: required
  validate-prd: optional
  create-design: conditional

Scan for existing workflow output files to auto-detect completion:

For each workflow in the list, check common output locations:

• {output_folder}/brainstorm-*.md for brainstorm-project
• {output_folder}/research-*.md for research
• {output_folder}/product-brief.md for product-brief
• {output_folder}/prd.md for prd
• {output_folder}/ux-design.md for create-design
• {output_folder}/architecture.md for create-architecture
• {output_folder}/tech-spec.md for tech-spec
• {output_folder}/sprint-status.yaml for sprint-planning

CRITICAL: If file exists, replace status with ONLY the file path - nothing else. Example: product-brief: docs/product-brief.md NOT: product-brief: "completed - docs/product-brief.md" or any other text.

workflow_items

Set generated date to current date generated

Prepare all template variables for workflow-status-template.yaml:

• generated: {current_date}
• project_name: {project_name}
• project_type: {project_type}
• project_level: {project_level}
• field_type: {field_type}
• workflow_path_file: {workflow_path_file}
• workflow_items: {workflow_items from step 5}

Display a preview of what will be created:

Show the first workflow in each phase and total count:

"Ready to create workflow status tracking:

• Phase 1 ({phase_1_workflow_count} workflows): Starting with {first_workflow_phase_1}
• Phase 2 ({phase_2_workflow_count} workflows): Starting with {first_workflow_phase_2}
• Phase 3 ({phase_3_workflow_count} workflows): Starting with {first_workflow_phase_3}
• Phase 4 (Implementation tracked separately in sprint-status.yaml)

{{#if detected_completed_workflows}} Found existing work: {{#each detected_files}}

• {{workflow_name}}: {{file_path}} {{/each}} {{/if}}"

Ready to create your workflow status file? (y/n)

Generate YAML from workflow-status-template.yaml with all variables Save status file to {output_folder}/bmm-workflow-status.yaml

Identify the first non-completed workflow in the list Look up that workflow's agent and command from the path file

✅ Workflow status file created at {output_folder}/bmm-workflow-status.yaml

Next Steps:

{{#if detected_completed_workflows}} You have {{detected_count}} workflow(s) already completed. Great progress! {{/if}}

Next Workflow: {{next_workflow_name}}

Agent: {{next_agent}}

Command: /bmad:bmm:workflows:{{next_workflow_id}}

{{#if next_agent !== 'pm'}} It is recommended to start a new chat and load the {{next_agent}} agent before running the next workflow. {{/if}}