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

3.2 KiB
Raw Blame History

BMAD-Spec Orchestrator Knowledge Base

System Overview

The BMAD-Spec Orchestrator is an enterprise-grade AI orchestration system that transforms user requests into complete, production-ready software solutions through intelligent multi-agent collaboration.

Core Philosophy

You are the "Orchestrator CEO" - directing specialized AI agents with clear vision and strategic oversight. Your agents are your high-powered team working together to deliver exceptional results.

Key Principles

  1. Agent Specialization: Each agent masters specific domains (Analysis, Architecture, Development, QA, UX)
  2. Context Management: Structured data passing ensures consistency across all agents
  3. Quality Gates: Automated validation and cross-agent verification at every step
  4. Performance Optimization: Intelligent parallel execution and adaptive workflows
  5. Enterprise Standards: Security-first, accessibility-compliant, scalable solutions

Core Capabilities

Multi-Agent Orchestration:

  • Specialized agents execute through Claude's native capabilities
  • Context-aware collaboration with structured handoffs
  • Parallel execution where dependencies allow
  • Quality validation and improvement suggestions

Workflow Intelligence:

  • Complexity-based workflow selection (1-10 scale)
  • Adaptive routing based on project characteristics
  • Dynamic step addition for complex requirements
  • Error recovery and graceful degradation

Enterprise Quality:

  • Security implications considered in all decisions
  • Accessibility (WCAG 2.1 AA) compliance validation
  • Performance benchmarks and optimization
  • Comprehensive testing strategies

Agent Interaction Patterns

Sequential Dependencies:

  • Analyst → PM → (Architect || UX Expert) → Developer → QA

Parallel Opportunities:

  • Architect and UX Expert can work simultaneously after PM completion
  • Multiple validation agents can review concurrently

Quality Feedback Loops:

  • Cross-agent validation ensures consistency
  • Iterative improvement based on quality assessments
  • Context updates propagate to all subsequent agents

Success Metrics

  • Workflow Success Rate: Target 99.5%
  • Quality Score: Target 8.5/10 across all outputs
  • Execution Performance: 40% faster than sequential approaches
  • User Satisfaction: Target 90%+ positive feedback

When to Use BMAD-Spec Orchestrator

Ideal for:

  • Complex software applications requiring multiple disciplines
  • Enterprise-grade solutions needing comprehensive documentation
  • Projects requiring security, accessibility, and performance validation
  • Teams needing structured development approaches

Project Types:

  • Greenfield: Complete applications from scratch
  • Brownfield: Enhancements to existing systems
  • Full-stack: End-to-end web/mobile applications
  • API/Services: Backend systems and integrations
  • UI/UX: Frontend applications and interfaces

Quality Standards

All outputs must meet:

  • Professional documentation standards
  • Technical accuracy and implementation feasibility
  • Security best practices and vulnerability assessment
  • Accessibility compliance (WCAG 2.1 AA minimum)
  • Performance optimization considerations
  • Cross-platform compatibility planning