Files
pig-farm-controller/bmad/bmb/workflows/module-brief/instructions.md
2025-11-01 19:22:39 +08:00

6.5 KiB

Module Brief Instructions

The workflow execution engine is governed by: {project-root}/bmad/core/tasks/workflow.xml You MUST have already loaded and processed: {project-root}/bmad/bmb/workflows/module-brief/workflow.yaml Communicate in {communication_language} throughout the module brief creation process

Ask the user which mode they prefer: 1. **Interactive Mode** - Work through each section collaboratively with detailed questions 2. **Express Mode** - Quick essential questions only 3. **YOLO Mode** (#yolo) - Generate complete draft based on minimal input

Check for available inputs:

  • Brainstorming results from previous sessions
  • Existing module ideas or notes
  • Similar modules for inspiration

If brainstorming results exist, offer to load and incorporate them

Ask the user to describe their module idea. Probe for: - What problem does this module solve? - Who would use this module? - What makes this module exciting or unique? - Any inspiring examples or similar tools?

If they're stuck, offer creative prompts:

  • "Imagine you're a [role], what tools would make your life easier?"
  • "What repetitive tasks could be automated with agents?"
  • "What domain expertise could be captured in workflows?"

module_vision

Based on the vision, work with user to define:

Module Code (kebab-case):

  • Suggest 2-3 options based on their description
  • Ensure it's memorable and descriptive

Module Name (friendly):

  • Creative, engaging name that captures the essence

Module Category:

  • Domain-Specific (legal, medical, finance)
  • Creative (writing, gaming, music)
  • Technical (devops, testing, architecture)
  • Business (project management, marketing)
  • Personal (productivity, learning)

Personality Theme (optional but fun!):

  • Should the module have a consistent personality across agents?
  • Star Trek crew? Fantasy party? Corporate team? Reality show cast?

module_identity

Help user envision their agent team

For each agent, capture:

  • Role: What's their specialty?
  • Personality: How do they communicate? (reference communication styles)
  • Key Capabilities: What can they do?
  • Signature Commands: 2-3 main commands

Suggest agent archetypes based on module type:

  • The Orchestrator (manages other agents)
  • The Specialist (deep expertise)
  • The Helper (utility functions)
  • The Creator (generates content)
  • The Analyzer (processes and evaluates)

agent_architecture

Map out the workflow landscape

Categorize workflows:

Core Workflows (2-3 essential ones):

  • The primary value-delivery workflows
  • What users will use most often

Feature Workflows (3-5 specialized):

  • Specific capabilities
  • Advanced features

Utility Workflows (1-3 supporting):

  • Setup, configuration
  • Maintenance, cleanup

For each workflow, define:

  • Purpose (one sentence)
  • Input → Process → Output
  • Complexity (simple/standard/complex)

workflow_ecosystem

Create usage scenarios to validate the design

Write 2-3 user stories: "As a [user type], I want to [goal], so that [outcome]"

Then walk through how they'd use the module:

  1. They load [agent]
  2. They run [command/workflow]
  3. They get [result]
  4. This helps them [achievement]

This validates the module makes sense end-to-end.

user_scenarios

Assess technical requirements:

Data Requirements:

  • What data/files does the module need?
  • Any external APIs or services?
  • Storage or state management needs?

Integration Points:

  • Other BMAD modules it might use
  • External tools or platforms
  • Import/export formats

Complexity Assessment:

  • Simple (standalone, no dependencies)
  • Standard (some integrations, moderate complexity)
  • Complex (multiple systems, advanced features)

technical_planning

Define what success looks like:

Module Success Criteria:

  • What indicates the module is working well?
  • How will users measure value?
  • What feedback mechanisms?

Quality Standards:

  • Performance expectations
  • Reliability requirements
  • User experience goals

success_metrics

Create a phased approach:

Phase 1 - MVP (Minimum Viable Module):

  • 1 primary agent
  • 2-3 core workflows
  • Basic functionality

Phase 2 - Enhancement:

  • Additional agents
  • More workflows
  • Refined features

Phase 3 - Polish:

  • Advanced features
  • Optimizations
  • Nice-to-haves

development_roadmap

If user wants to add special touches:

Easter Eggs:

  • Hidden commands or responses
  • Fun interactions between agents

Delighters:

  • Unexpected helpful features
  • Personality quirks
  • Creative responses

Module Lore:

  • Backstory for agents
  • Thematic elements
  • Consistent universe

creative_features

Identify potential challenges:

Technical Risks:

  • Complex integrations
  • Performance concerns
  • Dependency issues

Usability Risks:

  • Learning curve
  • Complexity creep
  • User confusion

Scope Risks:

  • Feature bloat
  • Timeline expansion
  • Resource constraints

For each risk, note mitigation strategy.

risk_assessment

Review all sections with {user_name} Ensure module brief is ready for create-module workflow

Would {user_name} like to:

  1. Proceed directly to create-module workflow
  2. Save and refine later
  3. Generate additional planning documents

Inform {user_name} in {communication_language} that this brief can be fed directly into create-module workflow

final_brief