mirror of
https://github.com/SuperClaude-Org/SuperClaude_Framework.git
synced 2025-12-29 16:16:08 +00:00
docs: Comprehensive documentation update for SuperClaude V4 Beta
Updated all root documentation to reflect V4 Beta capabilities: Root Documentation: - VERSION: Updated to 4.0.0-beta.1 - README.md: Complete rewrite with V4 features (21 commands, 13 agents, 6 MCP servers) - ARCHITECTURE_OVERVIEW.md: Updated for V4 Beta with correct counts and new features - CHANGELOG.md: Added comprehensive V4.0.0-beta.1 release section - ROADMAP.md: Added V4 Beta current status and updated future vision - CONTRIBUTING.md: Updated architecture, testing, and contribution guidelines - SECURITY.md: Added V4 security features and version support table - MANIFEST.in: Updated to include new V4 directories - pyproject.toml: Updated URLs and description for V4 Beta User Documentation: - commands-guide.md: Updated to 21 commands with new V4 commands - superclaude-user-guide.md: Comprehensive V4 Beta features documentation - flags-guide.md: Updated with new V4 flags and agent system - installation-guide.md: V4 Beta installation including hooks system - agents-guide.md: NEW - Complete guide for 13 specialized agents - personas-guide.md: Renamed to personas-guide-v3-legacy.md Key V4 Beta Features Documented: - 21 specialized commands (added: brainstorm, reflect, save, select-tool) - 13 domain expert agents replacing persona system - 6 MCP servers (added Morphllm and Serena) - 4 Behavioral Modes (Brainstorming, Introspection, Task Management, Token Efficiency) - Session Lifecycle with cross-session persistence - Redesigned Hooks System with Python integration - SuperClaude-Lite minimal implementation - Comprehensive Templates system All documentation maintains friendly, accessible tone while accurately reflecting V4 Beta's advanced capabilities. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
NomenAK
641
Docs/agents-guide.md
Normal file
641
Docs/agents-guide.md
Normal file
@@ -0,0 +1,641 @@
|
||||
# SuperClaude Agents Guide 🤖
|
||||
|
||||
## Overview
|
||||
|
||||
SuperClaude V4 Beta features 13 specialized domain expert agents that automatically activate based on your task context. These intelligent agents replace the previous persona system with more advanced, focused capabilities that provide expert-level assistance across all aspects of software development.
|
||||
|
||||
**The simple truth**: You don't need to pick agents or memorize what they do. SuperClaude automatically brings in the right experts for each situation!
|
||||
|
||||
**Here's what actually happens:**
|
||||
- You type `/analyze auth.js` → Security auditor automatically jumps in 🛡️
|
||||
- You work on React components → Frontend specialist often takes over 🎨
|
||||
- You debug performance issues → Performance optimizer often helps ⚡
|
||||
- You write documentation → Technical writer usually helps out ✍️
|
||||
|
||||
**It's like having a smart team** that knows when to jump in and help, without you managing who does what.
|
||||
|
||||
## 🚀 Just Try These (No Agent Knowledge Required)
|
||||
|
||||
```bash
|
||||
# These automatically activate the right experts:
|
||||
/sc:analyze payment-system/ # → Security + backend experts auto-activate
|
||||
/sc:build react-app/ # → Frontend specialist takes over
|
||||
/sc:improve slow-queries.sql # → Performance optimizer jumps in
|
||||
/sc:troubleshoot "auth failing" # → Root cause analyzer + security expert coordinate
|
||||
/sc:brainstorm "task manager app" # → Brainstorm-PRD agent guides discovery
|
||||
```
|
||||
|
||||
**See the pattern?** You focus on what you want to do, SuperClaude figures out who should help.
|
||||
|
||||
---
|
||||
|
||||
## The SuperClaude Agent Team 👥
|
||||
|
||||
### Core Development Agents 🔧
|
||||
|
||||
#### 🐍 `python-ultimate-expert` - Master Python Architect
|
||||
**What they do**: Production-ready Python development with SOLID principles, clean architecture, and comprehensive testing
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Python file detection (.py, requirements.txt, pyproject.toml)
|
||||
- Keywords: "python", "django", "fastapi", "flask", "asyncio"
|
||||
- Python-specific patterns and frameworks
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Production Standards**: SOLID principles, clean architecture, domain-driven design
|
||||
- **Testing Excellence**: TDD with 95%+ coverage, pytest, property-based testing
|
||||
- **Security First**: OWASP compliance, input validation, secrets management
|
||||
- **Performance Optimization**: Profiling, async programming, memory management
|
||||
- **Modern Tooling**: uv/poetry, black, ruff, mypy, pre-commit hooks
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:build python-api --focus security # → Production-ready FastAPI with security
|
||||
/sc:improve legacy-python/ --focus quality # → Refactor to SOLID principles
|
||||
/sc:test python-service/ --comprehensive # → Full test suite with coverage
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Context7**: Python framework documentation and best practices
|
||||
- **Sequential**: Complex architecture analysis and system design
|
||||
- **Magic**: Python web framework UI components
|
||||
|
||||
---
|
||||
|
||||
#### ⚙️ `backend-engineer` - Reliable Systems Specialist
|
||||
**What they do**: Server-side development, APIs, databases, and reliability engineering
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "API", "database", "server", "microservices", "reliability"
|
||||
- Backend file patterns (server.js, app.py, database schemas)
|
||||
- Infrastructure and service-related tasks
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Reliability First**: 99.9% uptime targets, fault tolerance, graceful degradation
|
||||
- **API Excellence**: RESTful design, GraphQL, comprehensive validation
|
||||
- **Database Mastery**: Schema design, query optimization, ACID compliance
|
||||
- **Security Implementation**: Authentication, authorization, zero trust architecture
|
||||
- **Observable Systems**: Structured logging, monitoring, alerting
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: 99.9% uptime with zero data loss tolerance
|
||||
- Secondary: <200ms API response time, comprehensive error handling
|
||||
- Success: Fault-tolerant systems meeting all reliability requirements
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:design user-management-api # → Reliable API with proper auth
|
||||
/sc:optimize database-queries/ # → Performance tuning and indexing
|
||||
/sc:implement payment-processing # → Secure, reliable payment system
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Context7**: Backend framework patterns and database documentation
|
||||
- **Sequential**: System architecture analysis and reliability planning
|
||||
- **Serena**: Cross-service dependency analysis and integration testing
|
||||
|
||||
---
|
||||
|
||||
#### 🎨 `frontend-specialist` - Accessible UI Expert
|
||||
**What they do**: User interface development with focus on accessibility, performance, and user experience
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "component", "responsive", "accessibility", "UI", "UX", "react", "vue"
|
||||
- Frontend file patterns (.jsx, .vue, .component.ts, CSS files)
|
||||
- User interface and design-related tasks
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Accessibility by Default**: WCAG 2.1 AA compliance, screen reader support
|
||||
- **Performance Budget**: Core Web Vitals optimization, bundle size management
|
||||
- **Responsive Design**: Mobile-first approach, progressive enhancement
|
||||
- **Modern Frameworks**: React, Vue, Angular with best practices
|
||||
- **Design Systems**: Component libraries, design tokens, consistent patterns
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: WCAG 2.1 AA compliance (100%) with Core Web Vitals in green zone
|
||||
- Secondary: <3s load time on 3G networks, zero accessibility errors
|
||||
- Success: Accessible, performant UI meeting all compliance standards
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:build dashboard-components/ # → Accessible React components
|
||||
/sc:improve --focus accessibility ui/ # → WCAG compliance and optimization
|
||||
/sc:optimize bundle-performance # → Core Web Vitals improvement
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Magic**: Advanced UI component generation and design system integration
|
||||
- **Context7**: Frontend framework documentation and accessibility patterns
|
||||
- **Playwright**: Cross-browser testing and visual regression detection
|
||||
|
||||
---
|
||||
|
||||
#### 🚀 `devops-engineer` - Infrastructure Automation Expert
|
||||
**What they do**: Infrastructure automation, deployment pipelines, monitoring, and reliability engineering
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "deploy", "infrastructure", "CI/CD", "docker", "kubernetes", "monitoring"
|
||||
- DevOps file patterns (Dockerfile, docker-compose.yml, .github/workflows)
|
||||
- Infrastructure and deployment-related tasks
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Infrastructure as Code**: Terraform, CloudFormation, automated provisioning
|
||||
- **CI/CD Excellence**: GitHub Actions, automated testing, deployment pipelines
|
||||
- **Container Orchestration**: Docker, Kubernetes, microservices deployment
|
||||
- **Monitoring & Observability**: Logging, metrics, alerting, performance tracking
|
||||
- **Security Operations**: Security scanning, compliance, vulnerability management
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Zero-downtime deployments with automated rollback capability
|
||||
- Secondary: Infrastructure as code, comprehensive monitoring coverage
|
||||
- Success: Fully automated, observable, secure deployment infrastructure
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:deploy production-app # → Zero-downtime deployment pipeline
|
||||
/sc:build monitoring-stack # → Comprehensive observability setup
|
||||
/sc:secure infrastructure/ # → Security hardening and compliance
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Context7**: DevOps tools documentation and best practices
|
||||
- **Sequential**: Infrastructure architecture analysis and optimization
|
||||
- **Playwright**: Deployment validation and smoke testing
|
||||
|
||||
### Analysis & Quality Agents 🔍
|
||||
|
||||
#### 🛡️ `security-auditor` - Threat Modeling Expert
|
||||
**What they do**: Security vulnerability identification, threat modeling, and compliance verification
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "security", "vulnerability", "auth", "compliance", "penetration", "audit"
|
||||
- Security-sensitive code patterns (authentication, data handling, API endpoints)
|
||||
- Security assessment and review requests
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Vulnerability Assessment**: OWASP Top 10, CWE compliance, penetration testing mindset
|
||||
- **Threat Modeling**: Attack vector analysis, security risk assessment
|
||||
- **Compliance Verification**: Industry standards, regulatory requirements
|
||||
- **Security Architecture**: Zero trust principles, defense in depth
|
||||
- **Remediation Guidance**: Specific fixes with security rationale
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Zero critical vulnerabilities in production with OWASP Top 10 compliance
|
||||
- Secondary: All findings include remediation steps, clear severity classifications
|
||||
- Success: Complete security assessment with actionable remediation plan
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:scan --focus security auth-system/ # → Comprehensive security audit
|
||||
/sc:analyze payment-flow --security # → Threat modeling and risk assessment
|
||||
/sc:improve --fix vulnerabilities api/ # → Security hardening and fixes
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Context7**: Security framework documentation and compliance standards
|
||||
- **Sequential**: Complex threat modeling and vulnerability chain analysis
|
||||
- **Playwright**: Security testing and penetration validation
|
||||
|
||||
---
|
||||
|
||||
#### ⚡ `performance-optimizer` - Bottleneck Detection Expert
|
||||
**What they do**: Performance analysis, optimization, and bottleneck identification across all system layers
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "performance", "optimization", "slow", "bottleneck", "latency", "memory"
|
||||
- Performance-related issues and optimization requests
|
||||
- System profiling and benchmarking tasks
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Bottleneck Identification**: Systematic performance profiling and analysis
|
||||
- **Optimization Strategy**: Database queries, application code, infrastructure tuning
|
||||
- **Performance Monitoring**: Metrics collection, alerting, continuous optimization
|
||||
- **Load Testing**: Capacity planning, stress testing, scalability analysis
|
||||
- **Resource Management**: Memory optimization, CPU utilization, I/O efficiency
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Measurable performance improvements with baseline metrics
|
||||
- Secondary: Comprehensive profiling, optimization rationale documentation
|
||||
- Success: Performance targets met with sustainable optimization strategies
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:analyze --focus performance slow-api/ # → Bottleneck identification and fixes
|
||||
/sc:optimize database-queries/ # → Query performance tuning
|
||||
/sc:benchmark application-performance # → Load testing and capacity planning
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Sequential**: Complex performance analysis and systematic optimization
|
||||
- **Context7**: Performance optimization patterns and best practices
|
||||
- **Playwright**: Performance testing and real-world load simulation
|
||||
|
||||
---
|
||||
|
||||
#### 🔍 `root-cause-analyzer` - Systematic Investigation Expert
|
||||
**What they do**: Debugging, root cause analysis, and evidence-based problem solving
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "debug", "investigate", "troubleshoot", "root cause", "analyze", "broken"
|
||||
- Bug reports, system failures, and complex problem investigation
|
||||
- Unknown system behavior and mysterious issues
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Systematic Investigation**: Evidence collection, hypothesis testing, pattern analysis
|
||||
- **Debug Methodology**: Step-by-step problem isolation and reproduction
|
||||
- **Root Cause Identification**: Deep analysis beyond surface symptoms
|
||||
- **Solution Validation**: Testing fixes and preventing regression
|
||||
- **Knowledge Transfer**: Documenting findings and prevention strategies
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Evidence-based conclusions with reproducible findings
|
||||
- Secondary: Systematic investigation methodology, complete analysis
|
||||
- Success: Root cause identified with validated solution and prevention
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:troubleshoot "payment processing fails randomly" # → Systematic investigation
|
||||
/sc:analyze mysterious-bug/ # → Evidence-based debugging
|
||||
/sc:investigate system-outage-logs/ # → Root cause analysis
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Sequential**: Complex problem decomposition and systematic analysis
|
||||
- **Serena**: Code analysis and symbol-level investigation
|
||||
- **Context7**: Debugging patterns and troubleshooting methodologies
|
||||
|
||||
---
|
||||
|
||||
#### 🧪 `qa-specialist` - Quality Assurance Expert
|
||||
**What they do**: Testing strategy, quality gates, edge case detection, and comprehensive validation
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "test", "quality", "validation", "QA", "edge case", "coverage"
|
||||
- Testing-related tasks and quality assurance requests
|
||||
- Quality gate implementation and validation processes
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Testing Strategy**: Comprehensive test planning, risk-based testing
|
||||
- **Quality Assurance**: Edge case identification, failure mode analysis
|
||||
- **Test Automation**: Unit, integration, and end-to-end testing frameworks
|
||||
- **Quality Gates**: Automated quality validation and compliance checking
|
||||
- **Risk Assessment**: Quality risk analysis and mitigation strategies
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Comprehensive test coverage with edge case validation
|
||||
- Secondary: Automated quality gates, risk-based testing priorities
|
||||
- Success: Quality assurance processes preventing defects in production
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:test --comprehensive user-service/ # → Full testing strategy and implementation
|
||||
/sc:validate --quality critical-features/ # → Quality gate implementation
|
||||
/sc:analyze --focus testing legacy-code/ # → Testing strategy for existing code
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Playwright**: End-to-end testing and cross-browser validation
|
||||
- **Sequential**: Complex testing strategy and quality analysis
|
||||
- **Context7**: Testing framework documentation and QA best practices
|
||||
|
||||
### Architecture & Code Quality Agents 🏗️
|
||||
|
||||
#### 🏗️ `system-architect` - Large-Scale Design Expert
|
||||
**What they do**: System architecture, design patterns, and scalability planning
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "architecture", "design", "scalability", "system", "patterns", "microservices"
|
||||
- Large-scale system design and architectural decisions
|
||||
- Cross-system integration and design pattern implementation
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **System Architecture**: Large-scale system design, microservices architecture
|
||||
- **Design Patterns**: Architectural patterns, design principle application
|
||||
- **Scalability Planning**: Performance at scale, distributed system design
|
||||
- **Integration Design**: Cross-system integration, API design, service mesh
|
||||
- **Technology Strategy**: Technology selection, architectural decision records
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Scalable, maintainable architecture supporting business requirements
|
||||
- Secondary: Design pattern adherence, comprehensive architectural documentation
|
||||
- Success: Architecture enabling development team productivity and system reliability
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:design microservices-architecture # → System architecture and service design
|
||||
/sc:analyze --focus architecture system/ # → Architectural review and improvement
|
||||
/sc:plan scalability-improvements # → Scaling strategy and implementation
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Sequential**: Complex architectural analysis and design validation
|
||||
- **Context7**: Architectural patterns and framework documentation
|
||||
- **Serena**: Cross-system analysis and dependency management
|
||||
|
||||
---
|
||||
|
||||
#### 🔄 `code-refactorer` - Clean Code Specialist
|
||||
**What they do**: Code quality improvement, technical debt management, and clean code practices
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "refactor", "cleanup", "quality", "technical debt", "maintainability"
|
||||
- Code improvement and cleanup requests
|
||||
- Legacy code modernization and quality enhancement
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Code Quality**: Clean code principles, SOLID design patterns
|
||||
- **Refactoring Strategy**: Safe refactoring, incremental improvement
|
||||
- **Technical Debt**: Debt identification, prioritization, systematic reduction
|
||||
- **Design Patterns**: Pattern application, code structure improvement
|
||||
- **Maintainability**: Code readability, documentation, consistency
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Improved code maintainability with reduced technical debt
|
||||
- Secondary: Design pattern adherence, comprehensive test coverage
|
||||
- Success: Clean, maintainable code supporting long-term development productivity
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:improve --focus quality legacy-module/ # → Comprehensive code quality improvement
|
||||
/sc:refactor --safe complex-functions/ # → Safe refactoring with test coverage
|
||||
/sc:cleanup --technical-debt codebase/ # → Systematic technical debt reduction
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Morphllm**: Pattern-based code transformations and intelligent refactoring
|
||||
- **Serena**: Symbol-level analysis and large-scale refactoring coordination
|
||||
- **Sequential**: Complex refactoring strategy and impact analysis
|
||||
|
||||
### Communication & Knowledge Agents 📚
|
||||
|
||||
#### ✍️ `technical-writer` - Documentation Excellence Expert
|
||||
**What they do**: Technical documentation, API references, user guides, and knowledge management
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "document", "README", "guide", "documentation", "API docs", "tutorial"
|
||||
- Documentation creation and improvement requests
|
||||
- Knowledge transfer and educational content needs
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Technical Documentation**: API documentation, user guides, technical specifications
|
||||
- **Audience Analysis**: Content tailored to different technical expertise levels
|
||||
- **Information Architecture**: Structured, navigable documentation systems
|
||||
- **Accessibility**: WCAG compliant documentation, inclusive design principles
|
||||
- **Content Strategy**: Documentation planning, maintenance, and lifecycle management
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Flesch Reading Score 60-70 with zero ambiguity in instructions
|
||||
- Secondary: WCAG 2.1 AA compliance, complete working code examples
|
||||
- Success: Documentation enabling successful task completion without external assistance
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:document api-endpoints/ # → Comprehensive API documentation
|
||||
/sc:write user-guide --audience beginner # → User-friendly tutorial and guides
|
||||
/sc:improve --docs project-documentation/ # → Documentation quality enhancement
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Context7**: Documentation patterns and technical writing best practices
|
||||
- **Serena**: Documentation persistence and cross-reference management
|
||||
- **Sequential**: Complex documentation planning and content strategy
|
||||
|
||||
---
|
||||
|
||||
#### 👨🏫 `code-educator` - Learning & Mentorship Expert
|
||||
**What they do**: Code explanation, educational content, and knowledge transfer facilitation
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Keywords: "explain", "learn", "understand", "teach", "tutorial", "example"
|
||||
- Educational content and learning-focused requests
|
||||
- Code explanation and mentorship needs
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Educational Content**: Clear, progressive learning materials and tutorials
|
||||
- **Code Explanation**: Complex concept breakdown with practical examples
|
||||
- **Mentorship**: Guided learning experiences and skill development
|
||||
- **Knowledge Transfer**: Best practices education and team learning facilitation
|
||||
- **Interactive Learning**: Hands-on examples and practical exercises
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Clear, progressive learning experiences appropriate to audience level
|
||||
- Secondary: Comprehensive examples, practical exercises, concept reinforcement
|
||||
- Success: Learners successfully apply concepts and develop practical skills
|
||||
|
||||
**Example use cases**:
|
||||
```bash
|
||||
/sc:explain complex-algorithm --educational # → Step-by-step learning guide
|
||||
/sc:teach react-patterns --beginner # → Progressive React tutorial
|
||||
/sc:mentor junior-developer --focus testing # → Personalized learning guidance
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Context7**: Educational resources and learning materials
|
||||
- **Sequential**: Complex concept breakdown and progressive learning design
|
||||
- **Magic**: Interactive examples and educational component creation
|
||||
|
||||
### Special Purpose Agents 🎯
|
||||
|
||||
#### 🧠 `brainstorm-PRD` - Requirements Discovery Expert
|
||||
**What they do**: Transform ambiguous project ideas into concrete specifications through structured brainstorming
|
||||
|
||||
**Auto-activation triggers**:
|
||||
- Ambiguous project requests ("I want to build something that...")
|
||||
- Exploration keywords: brainstorm, explore, discuss, figure out, not sure
|
||||
- PRD creation and requirements discovery needs
|
||||
- `/sc:brainstorm` command usage
|
||||
|
||||
**Specialized capabilities**:
|
||||
- **Requirements Discovery**: Socratic questioning, stakeholder analysis
|
||||
- **PRD Creation**: Comprehensive product requirement documentation
|
||||
- **Brainstorming Facilitation**: Creative exploration with practical constraints
|
||||
- **Specification Development**: Abstract concepts to concrete requirements
|
||||
- **Risk Assessment**: Early constraint and dependency identification
|
||||
|
||||
**Quality standards**:
|
||||
- Primary: Requirements are complete and unambiguous before project handoff
|
||||
- Secondary: All stakeholder perspectives integrated, feasibility validated
|
||||
- Success: Comprehensive PRD enabling downstream agent execution
|
||||
|
||||
**Integration workflow**:
|
||||
```bash
|
||||
/sc:brainstorm "task management app" # → Interactive discovery session
|
||||
# → Automatic handoff to brainstorm-PRD agent
|
||||
# → PRD generation with structured requirements
|
||||
# → Ready for /sc:workflow implementation
|
||||
```
|
||||
|
||||
**Integration with MCP servers**:
|
||||
- **Sequential**: Complex requirements analysis and systematic discovery
|
||||
- **Context7**: Industry patterns and requirement frameworks
|
||||
- **Serena**: Session persistence and cross-project learning
|
||||
|
||||
---
|
||||
|
||||
## Agent Coordination & Integration 🤝
|
||||
|
||||
### Automatic Agent Collaboration
|
||||
|
||||
Agents often work together automatically. Here are common collaboration patterns:
|
||||
|
||||
#### **Multi-Domain Projects**
|
||||
```bash
|
||||
/sc:build full-stack-app/
|
||||
# Auto-coordinates: backend-engineer + frontend-specialist + system-architect
|
||||
```
|
||||
|
||||
#### **Security-Focused Development**
|
||||
```bash
|
||||
/sc:analyze --focus security payment-system/
|
||||
# Auto-coordinates: security-auditor + backend-engineer + performance-optimizer
|
||||
```
|
||||
|
||||
#### **Quality Improvement**
|
||||
```bash
|
||||
/sc:improve --focus quality legacy-codebase/
|
||||
# Auto-coordinates: code-refactorer + qa-specialist + system-architect
|
||||
```
|
||||
|
||||
### Integration with MCP Servers
|
||||
|
||||
Each agent leverages specific MCP servers for enhanced capabilities:
|
||||
|
||||
- **Context7**: Documentation patterns, framework best practices, compliance standards
|
||||
- **Sequential**: Complex analysis, systematic problem solving, architectural planning
|
||||
- **Magic**: UI component generation, design system integration, modern frameworks
|
||||
- **Playwright**: Cross-browser testing, visual validation, performance testing
|
||||
- **Morphllm**: Intelligent code transformations, pattern application, optimization
|
||||
- **Serena**: Memory operations, cross-reference management, symbol-level analysis
|
||||
|
||||
### Integration with Commands and Modes
|
||||
|
||||
Agents seamlessly integrate with SuperClaude's command system:
|
||||
|
||||
```bash
|
||||
# Commands automatically select appropriate agents
|
||||
/sc:analyze → root-cause-analyzer or system-architect (context-dependent)
|
||||
/sc:build → frontend-specialist, backend-engineer, or python-ultimate-expert
|
||||
/sc:test → qa-specialist with domain-specific coordination
|
||||
/sc:brainstorm → brainstorm-PRD for requirements discovery
|
||||
/sc:document → technical-writer with audience-appropriate formatting
|
||||
```
|
||||
|
||||
**Mode Integration**:
|
||||
- **Brainstorming Mode**: Activates brainstorm-PRD for discovery sessions
|
||||
- **Task Management Mode**: Coordinates multiple agents across complex workflows
|
||||
- **Token Efficiency Mode**: Optimizes agent communication for resource constraints
|
||||
|
||||
## Quick Reference 📋
|
||||
|
||||
### Agent Selection Cheat Sheet
|
||||
|
||||
| Agent | Best For | Auto-Activates On | Example Use |
|
||||
|-------|----------|-------------------|-------------|
|
||||
| 🐍 python-ultimate-expert | Python development | .py files, Python frameworks | Production Python APIs |
|
||||
| ⚙️ backend-engineer | Server-side systems | APIs, databases, services | Reliable backend systems |
|
||||
| 🎨 frontend-specialist | User interfaces | UI components, accessibility | Accessible web interfaces |
|
||||
| 🚀 devops-engineer | Infrastructure | CI/CD, deployment, monitoring | Deployment automation |
|
||||
| 🛡️ security-auditor | Security analysis | Security keywords, auth code | Vulnerability assessment |
|
||||
| ⚡ performance-optimizer | Performance tuning | "slow", "bottleneck", profiling | System optimization |
|
||||
| 🔍 root-cause-analyzer | Problem solving | "debug", "investigate", bugs | Complex bug investigation |
|
||||
| 🧪 qa-specialist | Quality assurance | Testing, validation, QA | Comprehensive test strategy |
|
||||
| 🏗️ system-architect | System design | Architecture, scalability | Large-scale system design |
|
||||
| 🔄 code-refactorer | Code improvement | Refactoring, cleanup, quality | Clean code practices |
|
||||
| ✍️ technical-writer | Documentation | Documentation requests | API documentation |
|
||||
| 👨🏫 code-educator | Learning & teaching | "explain", "learn", tutorials | Educational content |
|
||||
| 🧠 brainstorm-PRD | Requirements discovery | Ambiguous projects, brainstorming | Project specification |
|
||||
|
||||
### Most Useful Agent Combinations
|
||||
|
||||
**Full-Stack Development**:
|
||||
```bash
|
||||
# Automatically coordinates backend + frontend + architecture
|
||||
/sc:build modern-web-app/
|
||||
```
|
||||
|
||||
**Security & Performance Review**:
|
||||
```bash
|
||||
# Coordinates security + performance + quality analysis
|
||||
/sc:analyze --comprehensive production-system/
|
||||
```
|
||||
|
||||
**Learning & Development**:
|
||||
```bash
|
||||
# Coordinates educator + technical writer + domain expert
|
||||
/sc:explain complex-system --educational
|
||||
```
|
||||
|
||||
**Project Discovery to Implementation**:
|
||||
```bash
|
||||
# Brainstorm → PRD → Architecture → Implementation
|
||||
/sc:brainstorm "e-commerce platform"
|
||||
# → Automatic handoff through agent coordination
|
||||
```
|
||||
|
||||
## Best Practices 💡
|
||||
|
||||
### Getting Started (The Simple Way)
|
||||
1. **Just use normal commands** - Agents auto-activate based on your needs
|
||||
2. **Trust the automation** - SuperClaude usually picks better experts than manual selection
|
||||
3. **Focus on your work** - Not on managing which agent helps you
|
||||
4. **Let coordination happen** - Multiple agents work together automatically
|
||||
|
||||
### Advanced Usage (When You Want Control)
|
||||
1. **Manual agent selection** - Use agent names in commands when you want specific expertise
|
||||
2. **Cross-domain perspectives** - Ask security agents about frontend code for different viewpoints
|
||||
3. **Learning mode** - Use code-educator for explanation-focused assistance
|
||||
4. **Quality focus** - Combine qa-specialist with domain experts for comprehensive quality
|
||||
|
||||
### Common Patterns
|
||||
|
||||
**For New Projects**:
|
||||
```bash
|
||||
/sc:brainstorm "your project idea" # → Requirements discovery
|
||||
# → Automatic PRD generation and handoff
|
||||
# → Ready for implementation workflow
|
||||
```
|
||||
|
||||
**For Existing Code**:
|
||||
```bash
|
||||
/sc:analyze existing-system/ # → Appropriate domain expert auto-selected
|
||||
/sc:improve --focus quality code/ # → Quality-focused agent coordination
|
||||
/sc:secure legacy-application/ # → Security-focused analysis and hardening
|
||||
```
|
||||
|
||||
**For Learning**:
|
||||
```bash
|
||||
/sc:explain complex-concept --educational # → Code educator with domain expert
|
||||
/sc:document api/ --audience beginner # → Technical writer with appropriate level
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Final Notes 📝
|
||||
|
||||
**The real truth about agents** 💯:
|
||||
- **Auto-activation works remarkably well** - Usually better than trying to pick experts yourself
|
||||
- **You can completely ignore agent details** and still get excellent expert assistance
|
||||
- **Agents exist to help you** - Not to create complexity you need to manage
|
||||
- **Learning happens naturally** through use, not through studying agent descriptions
|
||||
|
||||
**Don't feel overwhelmed by the team** 🧘♂️:
|
||||
- You don't need to know what each agent does
|
||||
- SuperClaude handles expert selection intelligently
|
||||
- The detailed descriptions above are for curiosity, not necessity
|
||||
- Focus on your work - the right experts will show up when needed
|
||||
|
||||
**When you might want to know about agents**:
|
||||
- **Curiosity** - "What would a security expert think about this frontend code?"
|
||||
- **Learning** - "How would different experts approach this problem?"
|
||||
- **Specialization** - "I specifically need Python architecture expertise"
|
||||
- **Quality focus** - "I want comprehensive quality analysis from multiple angles"
|
||||
|
||||
**Keep it simple** 🎯:
|
||||
- Use normal commands like `/analyze some-code/` and `/build my-app/`
|
||||
- Let the right experts automatically show up
|
||||
- Agent coordination is available when you want it, not because you need it
|
||||
- Focus on building great software - we'll handle the expertise coordination
|
||||
|
||||
---
|
||||
|
||||
*Behind this sophisticated team of 13 specialists, SuperClaude remains simple to use. Just start coding and the right experts show up when needed! 🚀*
|
||||
@@ -2,7 +2,7 @@
|
||||
|
||||
## 💡 Don't Overthink It - SuperClaude Tries to Help
|
||||
|
||||
**The truth about these 17 commands**: You don't need to memorize them. Just start with `/sc:analyze` or `/sc:implement` and see what happens!
|
||||
**The truth about these 21 commands**: You don't need to memorize them. Just start with `/sc:analyze` or `/sc:implement` and see what happens!
|
||||
|
||||
**Here's how it usually works:**
|
||||
- Type `/` in Claude Code → See available commands
|
||||
@@ -20,18 +20,20 @@
|
||||
```bash
|
||||
/sc:index # See what's available
|
||||
/sc:analyze src/ # Tries to analyze your code smartly
|
||||
/sc:brainstorm "app idea" # Interactive help for exploring ideas
|
||||
/sc:workflow feature-100-prd.md # Creates step-by-step implementation workflow from PRD
|
||||
/sc:implement user-auth # Creates features and components (replaces v2 /build)
|
||||
/sc:build # Attempts intelligent project building
|
||||
/sc:improve messy-file.js # Tries to clean up code
|
||||
/sc:troubleshoot "error" # Attempts to help with problems
|
||||
/sc:save --checkpoint # Save your work and progress
|
||||
```
|
||||
|
||||
**That's honestly enough to get started.** Everything else below is here when you get curious about what other tools are available. 🛠️
|
||||
|
||||
---
|
||||
|
||||
A practical guide to all 16 SuperClaude slash commands. We'll be honest about what works well and what's still rough around the edges.
|
||||
A practical guide to all 21 SuperClaude V4 Beta slash commands. We'll be honest about what works well and what's still rough around the edges.
|
||||
|
||||
## Quick Reference 📋
|
||||
|
||||
@@ -56,6 +58,10 @@ A practical guide to all 16 SuperClaude slash commands. We'll be honest about wh
|
||||
| `/sc:task` | Project management | Planning system | Long-term feature planning, task tracking |
|
||||
| `/sc:workflow` | Implementation planning | Workflow system | Creating step-by-step workflows from PRDs |
|
||||
| `/sc:index` | Command navigation | Help system | Finding the right command for your task |
|
||||
| `/sc:brainstorm` | Interactive discovery | Socratic dialogue | Requirements gathering, idea exploration |
|
||||
| `/sc:reflect` | Task validation | Serena intelligence | Progress checking, completion validation |
|
||||
| `/sc:save` | Session persistence | Checkpoint system | Context saving, session management |
|
||||
| `/sc:select-tool` | Tool selection | Intelligence routing | Complex operation optimization |
|
||||
|
||||
**Pro tip**: Just try the ones that sound useful. SuperClaude usually tries to activate helpful experts and tools for each situation! 🎯
|
||||
|
||||
@@ -634,7 +640,7 @@ A practical guide to all 16 SuperClaude slash commands. We'll be honest about wh
|
||||
|
||||
**Gotchas**:
|
||||
- Simple but useful for discovery
|
||||
- Better than trying to remember all 16 commands
|
||||
- Better than trying to remember all 21 commands
|
||||
|
||||
---
|
||||
|
||||
@@ -671,6 +677,146 @@ A practical guide to all 16 SuperClaude slash commands. We'll be honest about wh
|
||||
- More useful at project start than during development
|
||||
- Helps with onboarding but not a replacement for good docs
|
||||
|
||||
## Session & Intelligence Commands 🧠
|
||||
|
||||
### `/brainstorm` - Interactive Requirements Discovery
|
||||
**What it does**: Interactive Socratic dialogue for exploring ideas and discovering requirements.
|
||||
|
||||
**When to use it**:
|
||||
- Starting with vague project ideas ("I want to build something...")
|
||||
- Need help figuring out what to build
|
||||
- Exploring requirements for new features
|
||||
- Creative problem solving and ideation
|
||||
|
||||
**Basic syntax**:
|
||||
```bash
|
||||
/sc:brainstorm "mobile app idea" # Explore app concept
|
||||
/sc:brainstorm --depth deep startup # Deep exploration of startup idea
|
||||
/sc:brainstorm --focus business ecom # Business-focused e-commerce planning
|
||||
```
|
||||
|
||||
**Useful flags**:
|
||||
- `--depth normal|deep` - Exploration thoroughness
|
||||
- `--focus business|technical|user` - Conversation focus area
|
||||
- `--prd` - Generate Product Requirements Document after dialogue
|
||||
|
||||
**Real examples**:
|
||||
```bash
|
||||
/sc:brainstorm "task management app" --prd
|
||||
/sc:brainstorm --depth deep --focus technical "real-time chat system"
|
||||
/sc:brainstorm "improve user onboarding" --focus user
|
||||
```
|
||||
|
||||
**Gotchas**:
|
||||
- Works best when you're genuinely uncertain about requirements
|
||||
- Quality of output depends on engagement in the dialogue
|
||||
- Can take 10-15 minutes for thorough exploration
|
||||
|
||||
---
|
||||
|
||||
### `/reflect` - Task Reflection and Validation
|
||||
**What it does**: Uses Serena intelligence for task validation, progress analysis, and completion verification.
|
||||
|
||||
**When to use it**:
|
||||
- Checking if you're on the right track with a task
|
||||
- Validating task completion before marking done
|
||||
- Analyzing collected information during complex work
|
||||
- Getting intelligent feedback on progress
|
||||
|
||||
**Basic syntax**:
|
||||
```bash
|
||||
/sc:reflect --type task # Reflect on current task approach
|
||||
/sc:reflect --type completion # Validate task completion
|
||||
/sc:reflect --type session # Analyze session progress
|
||||
```
|
||||
|
||||
**Useful flags**:
|
||||
- `--type task|completion|session` - Type of reflection
|
||||
- `--analyze` - Deep analysis of current context
|
||||
- `--validate` - Validation-focused reflection
|
||||
|
||||
**Real examples**:
|
||||
```bash
|
||||
/sc:reflect --type completion "implemented user auth"
|
||||
/sc:reflect --analyze --type session
|
||||
/sc:reflect --type task --validate "refactoring approach"
|
||||
```
|
||||
|
||||
**Gotchas**:
|
||||
- Requires Serena MCP server to be available
|
||||
- Most effective when you provide context about what you're working on
|
||||
- Best used at natural stopping points in work
|
||||
|
||||
---
|
||||
|
||||
### `/save` - Session Persistence and Checkpointing
|
||||
**What it does**: Saves session context, creates checkpoints, and manages project memory through Serena.
|
||||
|
||||
**When to use it**:
|
||||
- Saving work progress before ending session
|
||||
- Creating checkpoints before risky operations
|
||||
- Preserving insights and decisions for future sessions
|
||||
- Managing long-term project context
|
||||
|
||||
**Basic syntax**:
|
||||
```bash
|
||||
/sc:save # Basic session save
|
||||
/sc:save --checkpoint # Create checkpoint with analysis
|
||||
/sc:save --type summary # Save with session summary
|
||||
```
|
||||
|
||||
**Useful flags**:
|
||||
- `--checkpoint` - Create detailed checkpoint
|
||||
- `--type session|summary|insights` - Save type
|
||||
- `--analyze` - Include session analysis in save
|
||||
|
||||
**Real examples**:
|
||||
```bash
|
||||
/sc:save --checkpoint "before major refactoring"
|
||||
/sc:save --type summary --analyze
|
||||
/sc:save "completed authentication implementation"
|
||||
```
|
||||
|
||||
**Gotchas**:
|
||||
- Requires Serena MCP server for full functionality
|
||||
- Checkpoint creation may take a moment for analysis
|
||||
- Most valuable for complex, multi-session projects
|
||||
|
||||
---
|
||||
|
||||
### `/select-tool` - Intelligent Tool Selection
|
||||
**What it does**: Analyzes complex operations and recommends optimal tool combinations and approaches.
|
||||
|
||||
**When to use it**:
|
||||
- Uncertain about best approach for complex tasks
|
||||
- Want to optimize tool selection for efficiency
|
||||
- Need guidance on MCP server coordination
|
||||
- Planning multi-step technical operations
|
||||
|
||||
**Basic syntax**:
|
||||
```bash
|
||||
/sc:select-tool "large codebase refactoring" # Get tool recommendations
|
||||
/sc:select-tool --analyze "performance audit" # Analyze optimal approach
|
||||
/sc:select-tool --context react-app "UI testing" # Context-aware selection
|
||||
```
|
||||
|
||||
**Useful flags**:
|
||||
- `--analyze` - Deep analysis of optimal approach
|
||||
- `--context <tech>` - Provide technical context
|
||||
- `--efficiency` - Focus on performance optimization
|
||||
|
||||
**Real examples**:
|
||||
```bash
|
||||
/sc:select-tool --analyze "migrate 100+ files to TypeScript"
|
||||
/sc:select-tool --context nodejs --efficiency "API performance testing"
|
||||
/sc:select-tool "cross-browser testing setup"
|
||||
```
|
||||
|
||||
**Gotchas**:
|
||||
- Recommendations are guidance, not requirements
|
||||
- Most valuable for complex, multi-faceted operations
|
||||
- Consider your specific project context when following recommendations
|
||||
|
||||
## Command Tips & Patterns 💡
|
||||
|
||||
### Effective Flag Combinations
|
||||
@@ -696,6 +842,14 @@ A practical guide to all 16 SuperClaude slash commands. We'll be honest about wh
|
||||
|
||||
### Common Workflows
|
||||
|
||||
**New Project Discovery** (V4 Beta):
|
||||
```bash
|
||||
/sc:brainstorm "project idea" --prd # Explore and define requirements
|
||||
/sc:load --deep --summary # Understand existing codebase
|
||||
/sc:workflow requirements.md # Create implementation plan
|
||||
/sc:save --checkpoint "project planning" # Save planning insights
|
||||
```
|
||||
|
||||
**New Project Onboarding**:
|
||||
```bash
|
||||
/sc:load --deep --summary
|
||||
@@ -709,6 +863,7 @@ A practical guide to all 16 SuperClaude slash commands. We'll be honest about wh
|
||||
/sc:troubleshoot "specific error message" --logs
|
||||
/sc:analyze --focus security
|
||||
/sc:test --type unit affected-component
|
||||
/sc:reflect --type completion "bug analysis"
|
||||
```
|
||||
|
||||
**Code Quality Improvement**:
|
||||
@@ -717,6 +872,7 @@ A practical guide to all 16 SuperClaude slash commands. We'll be honest about wh
|
||||
/sc:improve --preview src/
|
||||
/sc:cleanup --safe
|
||||
/sc:test --coverage
|
||||
/sc:reflect --type completion "quality improvements"
|
||||
```
|
||||
|
||||
**Pre-deployment Checklist**:
|
||||
@@ -725,6 +881,15 @@ A practical guide to all 16 SuperClaude slash commands. We'll be honest about wh
|
||||
/sc:analyze --focus security
|
||||
/sc:build --type prod --optimize
|
||||
/sc:git --smart-commit
|
||||
/sc:save --checkpoint "pre-deployment validation"
|
||||
```
|
||||
|
||||
**Complex Task Planning** (V4 Beta):
|
||||
```bash
|
||||
/sc:select-tool "migrate to microservices" # Get approach recommendations
|
||||
/sc:reflect --type task "migration strategy" # Validate approach
|
||||
/sc:workflow migration-plan.md # Create detailed workflow
|
||||
/sc:save "migration planning complete" # Preserve insights
|
||||
```
|
||||
|
||||
### Troubleshooting Command Issues
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# SuperClaude Flags User Guide 🏁
|
||||
# SuperClaude V4 Beta Flags User Guide 🏁
|
||||
|
||||
## 🤖 Most Flags Activate Automatically - Don't Stress About It!
|
||||
|
||||
@@ -7,7 +7,7 @@
|
||||
**Here's what actually happens:**
|
||||
- You type `/analyze auth.js`
|
||||
- SuperClaude detects it's security-related code
|
||||
- **Usually adds** `--persona-security`, `--focus security`, `--validate`
|
||||
- **Usually activates** security-auditor agent, `--focus security`, `--validate`
|
||||
- You often get expert security analysis without managing any flags
|
||||
|
||||
**When might you manually use flags?**
|
||||
@@ -23,10 +23,11 @@
|
||||
|
||||
```bash
|
||||
# These work great with zero flag knowledge:
|
||||
/sc:analyze src/ # Auto-picks the right analysis flags
|
||||
/sc:analyze src/ # Auto-activates appropriate expert agents
|
||||
/sc:build # Auto-optimizes based on your project
|
||||
/sc:improve messy-code.js # Auto-activates quality and safety flags
|
||||
/sc:troubleshoot "weird error" # Auto-activates debugging and analysis flags
|
||||
/sc:improve messy-code.js # Auto-activates quality agents and safety flags
|
||||
/sc:troubleshoot "weird error" # Auto-activates root-cause-analyzer and debugging flags
|
||||
/sc:brainstorm "my app idea" # Auto-activates brainstorm-PRD agent for requirements
|
||||
```
|
||||
|
||||
**See? No flags needed.** Everything below is for when you get curious about what's happening behind the scenes.
|
||||
@@ -153,9 +154,21 @@ Enable specialized capabilities through MCP servers.
|
||||
#### `--play` / `--playwright`
|
||||
**What it does**: Enables Playwright for browser automation and testing
|
||||
**When to use**: E2E testing, performance monitoring
|
||||
**Auto-activates**: Test workflows, QA persona
|
||||
**Auto-activates**: Test workflows, QA specialist agent
|
||||
**Example**: `/test e2e --play`
|
||||
|
||||
#### `--morph` / `--fast-apply`
|
||||
**What it does**: Enables Morphllm for intelligent file editing with Fast Apply
|
||||
**When to use**: Multi-file edits, pattern-based transformations
|
||||
**Auto-activates**: Complex refactoring, bulk file updates
|
||||
**Example**: `/refactor --morph pattern-updates/`
|
||||
|
||||
#### `--serena` / `--semantic`
|
||||
**What it does**: Enables Serena for semantic analysis and memory operations
|
||||
**When to use**: Symbol analysis, project-wide context, session management
|
||||
**Auto-activates**: Complex symbol operations, memory commands
|
||||
**Example**: `/analyze --serena --semantic large-codebase/`
|
||||
|
||||
#### `--all-mcp`
|
||||
**What it does**: Enables all MCP servers simultaneously
|
||||
**When to use**: Complex multi-domain problems
|
||||
@@ -167,7 +180,7 @@ Enable specialized capabilities through MCP servers.
|
||||
**When to use**: Faster execution, don't need specialized features
|
||||
**Example**: `/analyze simple-script.js --no-mcp`
|
||||
|
||||
**💡 Tip**: MCP servers add capabilities but use more tokens. `--c7` for docs, `--seq` for thinking, `--magic` for UI.
|
||||
**💡 Tip**: MCP servers add capabilities but use more tokens. `--c7` for docs, `--seq` for thinking, `--magic` for UI, `--serena` for memory & semantic analysis.
|
||||
|
||||
---
|
||||
|
||||
@@ -198,6 +211,16 @@ For complex operations and workflows.
|
||||
**Auto-activates**: Polish, refine, enhance keywords
|
||||
**Example**: `/improve messy-code.js --loop`
|
||||
|
||||
#### `--iterations [n]`
|
||||
**What it does**: Control number of improvement cycles (1-10)
|
||||
**When to use**: Controlling iterative enhancement depth
|
||||
**Example**: `/improve --loop --iterations 5`
|
||||
|
||||
#### `--interactive`
|
||||
**What it does**: Enables user confirmation between iterations
|
||||
**When to use**: Want control over iterative process
|
||||
**Example**: `/improve --loop --interactive`
|
||||
|
||||
#### `--concurrency [n]`
|
||||
**What it does**: Control max concurrent sub-agents (1-15)
|
||||
**When to use**: Controlling resource usage
|
||||
@@ -207,6 +230,31 @@ For complex operations and workflows.
|
||||
|
||||
---
|
||||
|
||||
### Session & Workflow Flags 🔄
|
||||
|
||||
These V4 Beta flags control session management and advanced workflows.
|
||||
|
||||
#### `--brainstorm`
|
||||
**What it does**: Activates brainstorming mode for requirements discovery
|
||||
**When to use**: Ambiguous project ideas, need specification development
|
||||
**Auto-activates**: Exploration keywords, "not sure", project planning
|
||||
**Example**: `/brainstorm "task management app" --brainstorm`
|
||||
|
||||
#### `--max-rounds [n]`
|
||||
**What it does**: Controls brainstorming dialogue rounds (default: 15)
|
||||
**When to use**: Controlling brainstorming session depth
|
||||
**Example**: `/brainstorm idea --max-rounds 10`
|
||||
|
||||
#### `--prd`
|
||||
**What it does**: Generates Product Requirements Document from brainstorming
|
||||
**When to use**: Convert discussions to formal specifications
|
||||
**Auto-activates**: End of successful brainstorming sessions
|
||||
**Example**: `/brainstorm app-idea --prd`
|
||||
|
||||
**💡 Tip**: V4 Beta introduces powerful session management - let brainstorming mode guide requirement discovery naturally.
|
||||
|
||||
---
|
||||
|
||||
### Focus & Scope Flags 🎯
|
||||
|
||||
Direct SuperClaude's attention to specific areas.
|
||||
@@ -221,12 +269,12 @@ Direct SuperClaude's attention to specific areas.
|
||||
**What it does**: Focuses analysis on specific domain
|
||||
**Example**: `/analyze --focus security --scope project`
|
||||
|
||||
#### Persona Flags
|
||||
**Available personas**: architect, frontend, backend, analyzer, security, mentor, refactorer, performance, qa, devops, scribe
|
||||
**What they do**: Activates specialist behavior patterns
|
||||
**Example**: `/analyze --persona-security` - Security-focused analysis
|
||||
#### Agent Flags
|
||||
**Available agents**: system-architect, frontend-specialist, backend-engineer, root-cause-analyzer, security-auditor, code-educator, code-refactorer, performance-optimizer, qa-specialist, devops-engineer, technical-writer, python-ultimate-expert, brainstorm-PRD
|
||||
**What they do**: Activates specialized domain expert agents (V4 Beta enhancement)
|
||||
**Example**: `/analyze --agent security-auditor` - Security-focused analysis with expert agent
|
||||
|
||||
**💡 Tip**: `--focus` is great for targeted analysis. Personas auto-activate but manual control helps.
|
||||
**💡 Tip**: `--focus` is great for targeted analysis. Expert agents auto-activate but manual control helps.
|
||||
|
||||
---
|
||||
|
||||
@@ -330,11 +378,11 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
### Domain-Based
|
||||
```bash
|
||||
/sc:build react-app/
|
||||
# Auto-adds: --c7 --persona-frontend
|
||||
# Auto-activates: frontend-specialist agent + --c7
|
||||
# Why: Frontend framework detected
|
||||
|
||||
/sc:analyze --focus security
|
||||
# Auto-adds: --persona-security --validate
|
||||
# Auto-activates: security-auditor agent + --validate
|
||||
# Why: Security focus triggers security specialist
|
||||
```
|
||||
|
||||
@@ -357,20 +405,20 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
|
||||
**Comprehensive Code Review**:
|
||||
```bash
|
||||
/sc:review codebase/ --persona-qa --think-hard --focus quality --validate --c7
|
||||
# → QA specialist + deep thinking + quality focus + validation + docs
|
||||
/sc:review codebase/ --agent qa-specialist --think-hard --focus quality --validate --c7
|
||||
# → QA specialist agent + deep thinking + quality focus + validation + docs
|
||||
```
|
||||
|
||||
**Legacy System Modernization**:
|
||||
```bash
|
||||
/sc:improve legacy/ --wave-mode force --persona-architect --safe-mode --loop --c7
|
||||
# → Wave orchestration + architect perspective + safety + iteration + docs
|
||||
/sc:improve legacy/ --wave-mode force --agent system-architect --safe-mode --loop --c7
|
||||
# → Wave orchestration + architect agent + safety + iteration + docs
|
||||
```
|
||||
|
||||
**Security Audit**:
|
||||
```bash
|
||||
/sc:scan --persona-security --ultrathink --focus security --validate --seq
|
||||
# → Security specialist + maximum thinking + security focus + validation + systematic analysis
|
||||
/sc:scan --agent security-auditor --ultrathink --focus security --validate --seq
|
||||
# → Security auditor agent + maximum thinking + security focus + validation + systematic analysis
|
||||
```
|
||||
|
||||
### Performance Optimization
|
||||
@@ -392,16 +440,16 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
**Bug Investigation Workflow**:
|
||||
```bash
|
||||
/sc:troubleshoot "specific error" --seq --think --validate
|
||||
/sc:analyze affected-files/ --focus quality --persona-analyzer
|
||||
/sc:analyze affected-files/ --focus quality --agent root-cause-analyzer
|
||||
/sc:test --play --coverage
|
||||
```
|
||||
|
||||
**Feature Development Workflow**:
|
||||
```bash
|
||||
/sc:design new-feature --persona-architect --c7
|
||||
/sc:build --magic --persona-frontend --validate
|
||||
/sc:design new-feature --agent system-architect --c7
|
||||
/sc:build --magic --agent frontend-specialist --validate
|
||||
/sc:test --play --coverage
|
||||
/sc:document --persona-scribe --c7
|
||||
/sc:document --agent technical-writer --c7
|
||||
```
|
||||
|
||||
## Quick Reference 📋
|
||||
@@ -417,6 +465,8 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
| `--focus security` | Security focus | Security concerns |
|
||||
| `--delegate auto` | Parallel processing | Large codebases |
|
||||
| `--validate` | Check before action | Risky operations |
|
||||
| `--brainstorm` | Requirements discovery | Ambiguous projects |
|
||||
| `--serena` | Semantic analysis | Symbol operations |
|
||||
|
||||
### Flag Combinations That Work Well
|
||||
```bash
|
||||
@@ -430,13 +480,16 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
--delegate auto --uc --focus
|
||||
|
||||
# Learning
|
||||
--verbose --c7 --persona-mentor
|
||||
--verbose --c7 --agent code-educator
|
||||
|
||||
# Security work
|
||||
--persona-security --focus security --validate
|
||||
--agent security-auditor --focus security --validate
|
||||
|
||||
# Performance work
|
||||
--persona-performance --focus performance --play
|
||||
--agent performance-optimizer --focus performance --play
|
||||
|
||||
# Requirements discovery
|
||||
--brainstorm --max-rounds 10 --prd
|
||||
```
|
||||
|
||||
### Auto-Activation Triggers
|
||||
@@ -446,16 +499,18 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
- **--delegate**: >7 directories or >50 files
|
||||
- **--c7**: Framework imports, documentation requests
|
||||
- **--seq**: Debugging keywords, --think flags
|
||||
- **Personas**: Domain-specific keywords and patterns
|
||||
- **--serena**: Symbol operations, memory commands
|
||||
- **--brainstorm**: Exploration keywords, ambiguous requests
|
||||
- **Expert Agents**: Domain-specific keywords and patterns
|
||||
|
||||
## Troubleshooting Flag Issues 🚨
|
||||
|
||||
### Common Problems
|
||||
|
||||
**"Flags don't seem to work"**
|
||||
- Check spelling (common typos: `--ultracompresed`, `--persona-fronted`)
|
||||
- Check spelling (common typos: `--ultracompresed`, `--agent fronted-specialist`)
|
||||
- Some flags need values: `--scope project`, `--focus security`
|
||||
- Flag conflicts: `--no-mcp` overrides `--c7`, `--seq`, etc.
|
||||
- Flag conflicts: `--no-mcp` overrides `--c7`, `--seq`, `--serena`, etc.
|
||||
|
||||
**"Operation too slow"**
|
||||
- Try `--uc` for compression
|
||||
@@ -470,7 +525,7 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
**"Not thorough enough"**
|
||||
- Add `--think` or `--think-hard`
|
||||
- Enable relevant MCP servers: `--seq`, `--c7`
|
||||
- Use appropriate persona: `--persona-analyzer`
|
||||
- Use appropriate agent: `--agent root-cause-analyzer`
|
||||
|
||||
**"Changes too risky"**
|
||||
- Always use `--safe-mode` for important code
|
||||
@@ -480,9 +535,9 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
### Flag Conflicts
|
||||
|
||||
**These override others**:
|
||||
- `--no-mcp` overrides all MCP flags (`--c7`, `--seq`, etc.)
|
||||
- `--no-mcp` overrides all MCP flags (`--c7`, `--seq`, `--serena`, etc.)
|
||||
- `--safe-mode` overrides optimization flags
|
||||
- Last persona flag wins: `--persona-frontend --persona-backend` → backend
|
||||
- Last agent flag wins: `--agent frontend-specialist --agent backend-engineer` → backend-engineer
|
||||
|
||||
**Precedence order**:
|
||||
1. Safety flags (`--safe-mode`) beat optimization
|
||||
@@ -499,8 +554,8 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
4. **Trust the automation** - SuperClaude usually picks reasonable defaults
|
||||
|
||||
### Getting Advanced (If You Want To)
|
||||
1. **Experiment with overrides** - Try `--persona-security` on non-security code for different perspectives
|
||||
2. **Learn the useful combos** - `--safe-mode --validate` for important stuff
|
||||
1. **Experiment with overrides** - Try `--agent security-auditor` on non-security code for different perspectives
|
||||
2. **Learn the useful combos** - `--safe-mode --validate` for important stuff, `--brainstorm --prd` for new projects
|
||||
3. **Understand the performance trade-offs** - Fast (`--uc --no-mcp`) vs thorough (`--think-hard --all-mcp`)
|
||||
4. **Use flags for learning** - `--verbose` when you want to understand what's happening
|
||||
|
||||
@@ -508,7 +563,8 @@ SuperClaude usually adds flags based on context. Here's when it tries:
|
||||
- **For speed**: `--uc --no-mcp --scope file`
|
||||
- **For thoroughness**: `--think-hard --all-mcp --delegate auto`
|
||||
- **For safety**: `--safe-mode --validate --preview`
|
||||
- **For learning**: `--verbose --c7 --persona-mentor`
|
||||
- **For learning**: `--verbose --c7 --agent code-educator`
|
||||
- **For project discovery**: `--brainstorm --max-rounds 15 --prd`
|
||||
|
||||
---
|
||||
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
# SuperClaude Installation Guide 📦
|
||||
# SuperClaude V4 Beta Installation Guide 📦
|
||||
|
||||
## 🎯 It's Easier Than It Looks!
|
||||
|
||||
@@ -13,13 +13,13 @@ uv add SuperClaude
|
||||
|
||||
**Option B: From Source**
|
||||
```bash
|
||||
git clone https://github.com/NomenAK/SuperClaude.git
|
||||
cd SuperClaude
|
||||
git clone https://github.com/SuperClaude-Org/SuperClaude_Framework.git
|
||||
cd SuperClaude_Framework
|
||||
uv sync
|
||||
```
|
||||
### 🔧 UV / UVX Setup Guide
|
||||
|
||||
SuperClaude v3 also supports installation via [`uv`](https://github.com/astral-sh/uv) (a faster, modern Python package manager) or `uvx` for cross-platform usage.
|
||||
SuperClaude V4 Beta supports installation via [`uv`](https://github.com/astral-sh/uv) (a faster, modern Python package manager) or `uvx` for cross-platform usage.
|
||||
|
||||
### 🌀 Install with `uv`
|
||||
|
||||
@@ -48,7 +48,7 @@ uvx pip install SuperClaude
|
||||
```
|
||||
## 🔧 UV / UVX Setup Guide
|
||||
|
||||
SuperClaude v3 also supports installation via [`uv`](https://github.com/astral-sh/uv) (a faster, modern Python package manager) or `uvx` for cross-platform usage.
|
||||
SuperClaude V4 Beta supports installation via [`uv`](https://github.com/astral-sh/uv) (a faster, modern Python package manager) or `uvx` for cross-platform usage.
|
||||
|
||||
### 🌀 Install with `uv`
|
||||
|
||||
@@ -106,7 +106,7 @@ SuperClaude install
|
||||
|
||||
---
|
||||
|
||||
A comprehensive guide to installing SuperClaude v3. But remember - most people never need to read past the quick start above! 😊
|
||||
A comprehensive guide to installing SuperClaude V4 Beta. But remember - most people never need to read past the quick start above! 😊
|
||||
|
||||
## Before You Start 🔍
|
||||
|
||||
@@ -169,8 +169,10 @@ SuperClaude install --quick
|
||||
, `python3 -m SuperClaude commands` or also `python3 SuperClaude commands`**
|
||||
|
||||
**What you just got:**
|
||||
- ✅ All 16 smart commands that auto-activate experts
|
||||
- ✅ 11 specialist personas that know when to help
|
||||
- ✅ All 16 smart commands that auto-activate experts
|
||||
- ✅ 13 specialized domain expert agents that know when to help
|
||||
- ✅ Advanced session management with brainstorming mode
|
||||
- ✅ Python hooks system for intelligent framework coordination
|
||||
- ✅ Intelligent routing that figures out complexity for you
|
||||
- ✅ About 2 minutes of your time and ~50MB disk space
|
||||
|
||||
@@ -205,15 +207,25 @@ SuperClaude install --quick
|
||||
- **Good for**: Most users, general development
|
||||
- **Includes**: Everything in minimal + specialized commands like `/analyze`, `/build`, `/improve`
|
||||
|
||||
### 🪶 SuperClaude-Lite Installation (V4 Beta)
|
||||
```bash
|
||||
SuperClaude install --lite
|
||||
```
|
||||
- **What**: Streamlined version with core features only
|
||||
- **Time**: ~1 minute
|
||||
- **Space**: ~25MB
|
||||
- **Good for**: Resource-constrained environments, basic enhancement
|
||||
- **Includes**: Essential framework features without advanced orchestration
|
||||
|
||||
### 🔧 Developer Installation
|
||||
```bash
|
||||
SuperClaude install --profile developer
|
||||
```
|
||||
- **What**: Everything including MCP server integration
|
||||
- **What**: Everything including MCP server integration + Hooks system
|
||||
- **Time**: ~5 minutes
|
||||
- **Space**: ~100MB
|
||||
- **Good for**: Power users, contributors, advanced workflows
|
||||
- **Includes**: Everything + Context7, Sequential, Magic, Playwright servers
|
||||
- **Includes**: Everything + Context7, Sequential, Magic, Playwright, Morphllm, Serena servers + Python hooks
|
||||
|
||||
### 🎛️ Interactive Installation
|
||||
```bash
|
||||
@@ -278,8 +290,8 @@ pip install .
|
||||
|
||||
**Option 3: Clone from Git**
|
||||
```bash
|
||||
git clone <repository-url>
|
||||
cd SuperClaude
|
||||
git clone https://github.com/SuperClaude-Org/SuperClaude_Framework.git
|
||||
cd SuperClaude_Framework
|
||||
pip install .
|
||||
```
|
||||
|
||||
@@ -316,8 +328,9 @@ Here's what happens when you install:
|
||||
3. **Core Files** - Copies framework documentation files
|
||||
4. **Commands** - Installs slash command definitions (if selected)
|
||||
5. **MCP Servers** - Downloads and configures MCP servers (if selected)
|
||||
6. **Configuration** - Sets up `settings.json` with your preferences
|
||||
7. **Validation** - Tests that everything works
|
||||
6. **Hooks System** - Installs Python hooks for framework coordination (developer profile)
|
||||
7. **Configuration** - Sets up `settings.json` with your preferences
|
||||
8. **Validation** - Tests that everything works
|
||||
|
||||
The installer shows progress and will tell you if anything goes wrong.
|
||||
|
||||
@@ -341,19 +354,18 @@ ls ~/.claude/
|
||||
|
||||
### What Got Installed 📂
|
||||
|
||||
SuperClaude installs to `~/.claude/` by default. Here's what you'll find:
|
||||
SuperClaude V4 Beta installs to `~/.claude/` by default. Here's what you'll find:
|
||||
|
||||
```
|
||||
~/.claude/
|
||||
├── CLAUDE.md # Main framework entry point
|
||||
├── COMMANDS.md # Available slash commands
|
||||
├── FLAGS.md # Command flags and options
|
||||
├── PERSONAS.md # Smart persona system
|
||||
├── PRINCIPLES.md # Development principles
|
||||
├── RULES.md # Operational rules
|
||||
├── MCP.md # MCP server integration
|
||||
├── MODES.md # Operational modes
|
||||
├── ORCHESTRATOR.md # Intelligent routing
|
||||
├── MCP_*.md # MCP server configurations
|
||||
├── MODE_*.md # Operational modes
|
||||
├── SESSION_LIFECYCLE.md # Session management
|
||||
├── settings.json # Configuration file
|
||||
└── commands/ # Individual command definitions
|
||||
├── analyze.md
|
||||
@@ -362,10 +374,23 @@ SuperClaude installs to `~/.claude/` by default. Here's what you'll find:
|
||||
└── ... (13 more)
|
||||
```
|
||||
|
||||
**For Developer Installation (+Hooks System)**:
|
||||
```
|
||||
~/.claude/
|
||||
├── [above files] # Core framework files
|
||||
└── hooks/ # Python hooks system (developer profile)
|
||||
├── framework_coordinator/
|
||||
├── session_lifecycle/
|
||||
├── performance_monitor/
|
||||
├── quality_gates/
|
||||
└── install_hooks.py # Hook installation script
|
||||
```
|
||||
|
||||
**What each file does:**
|
||||
- **CLAUDE.md** - Tells Claude Code about SuperClaude and loads other files
|
||||
- **settings.json** - Configuration (MCP servers, hooks, etc.)
|
||||
- **commands/** - Detailed definitions for each slash command
|
||||
- **hooks/** - Python hooks for advanced framework coordination (developer only)
|
||||
|
||||
### First Steps 🎯
|
||||
|
||||
@@ -377,10 +402,39 @@ Try these commands to get started:
|
||||
/sc:analyze README.md # Analyze a file
|
||||
/sc:build --help # See build options
|
||||
/sc:improve --help # See improvement options
|
||||
/sc:brainstorm "my app idea" # Try V4 Beta brainstorming mode
|
||||
```
|
||||
|
||||
**Don't worry if it seems overwhelming** - SuperClaude enhances Claude Code gradually. You can use as much or as little as you want.
|
||||
|
||||
### Installing Hooks System (V4 Beta) 🔗
|
||||
|
||||
If you want to add the Python hooks system to an existing installation:
|
||||
|
||||
```bash
|
||||
# Install hooks to existing SuperClaude installation
|
||||
cd SuperClaude_Framework
|
||||
python3 SuperClaude/Hooks/scripts/install_hooks.py
|
||||
|
||||
# Or upgrade existing installation with hooks
|
||||
SuperClaude install --upgrade --hooks
|
||||
```
|
||||
|
||||
**What the hooks system provides:**
|
||||
- **Framework Coordinator**: Intelligent MCP server suggestions
|
||||
- **Session Lifecycle**: Automatic checkpoint triggers
|
||||
- **Performance Monitor**: Real-time performance tracking (<100ms targets)
|
||||
- **Quality Gates**: 8-step validation system
|
||||
|
||||
**Verification:**
|
||||
```bash
|
||||
# Test hooks installation
|
||||
python3 SuperClaude/Hooks/scripts/test_hooks.py
|
||||
|
||||
# Check hook status
|
||||
SuperClaude hooks --status
|
||||
```
|
||||
|
||||
## Managing Your Installation 🛠️
|
||||
|
||||
### Updates 📅
|
||||
@@ -516,7 +570,7 @@ SuperClaude install --quick --dry-run
|
||||
### Still Having Issues? 🤔
|
||||
|
||||
**Check our troubleshooting resources:**
|
||||
- GitHub Issues: https://github.com/NomenAK/SuperClaude/issues
|
||||
- GitHub Issues: https://github.com/SuperClaude-Org/SuperClaude_Framework/issues
|
||||
- Look for existing issues similar to yours
|
||||
- Create a new issue if you can't find a solution
|
||||
|
||||
@@ -606,4 +660,4 @@ Thanks for trying SuperClaude! We hope it makes your development workflow a bit
|
||||
|
||||
---
|
||||
|
||||
*Last updated: July 2024 - Let us know if anything in this guide is wrong or confusing!*
|
||||
*Last updated: January 2025 (V4 Beta) - Let us know if anything in this guide is wrong or confusing!*
|
||||
|
||||
@@ -22,21 +22,22 @@ The detailed guides below? They're here **when you want to understand** what jus
|
||||
|
||||
---
|
||||
|
||||
A comprehensive guide to understanding and using SuperClaude v3.0 effectively. But remember - you can skip straight to trying it out!
|
||||
A comprehensive guide to understanding and using SuperClaude V4 Beta effectively. But remember - you can skip straight to trying it out!
|
||||
|
||||
## Table of Contents 📖
|
||||
|
||||
1. [Welcome & Overview](#welcome--overview-)
|
||||
2. [Core Components](#core-components-)
|
||||
3. [The Three Operational Modes](#the-three-operational-modes-)
|
||||
4. [The Orchestrator System](#the-orchestrator-system-)
|
||||
5. [Rules & Principles](#rules--principles-)
|
||||
6. [Getting Started Workflows](#getting-started-workflows-)
|
||||
7. [Integration & Coordination](#integration--coordination-)
|
||||
8. [Practical Examples](#practical-examples-)
|
||||
9. [Tips & Best Practices](#tips--best-practices-)
|
||||
10. [Troubleshooting](#troubleshooting--common-issues-)
|
||||
11. [What's Next](#whats-next-)
|
||||
3. [The Four Behavioral Modes](#the-four-behavioral-modes-)
|
||||
4. [Session Lifecycle System](#session-lifecycle-system-)
|
||||
5. [The Orchestrator System](#the-orchestrator-system-)
|
||||
6. [Rules & Principles](#rules--principles-)
|
||||
7. [Getting Started Workflows](#getting-started-workflows-)
|
||||
8. [Integration & Coordination](#integration--coordination-)
|
||||
9. [Practical Examples](#practical-examples-)
|
||||
10. [Tips & Best Practices](#tips--best-practices-)
|
||||
11. [Troubleshooting](#troubleshooting--common-issues-)
|
||||
12. [What's Next](#whats-next-)
|
||||
|
||||
---
|
||||
|
||||
@@ -46,19 +47,22 @@ A comprehensive guide to understanding and using SuperClaude v3.0 effectively. B
|
||||
|
||||
```bash
|
||||
# Try these commands in Claude Code:
|
||||
/sc:help # See what's available
|
||||
/sc:load # Initialize session with project context (NEW!)
|
||||
/sc:analyze README.md # SuperClaude analyzes your project
|
||||
/sc:workflow feature-prd.md # Generate implementation workflow from PRD (NEW!)
|
||||
/sc:implement user-auth # Create features and components (NEW in v3!)
|
||||
/sc:brainstorm "task app" # Interactive requirements discovery (NEW!)
|
||||
/sc:implement user-auth # Create features and components
|
||||
/sc:build # Smart build with auto-optimization
|
||||
/sc:improve messy-file.js # Clean up code automatically
|
||||
/sc:save # Save session state and insights (NEW!)
|
||||
```
|
||||
|
||||
**What just happened?** SuperClaude automatically:
|
||||
- Initialized persistent session context 🧠
|
||||
- Picked the right tools for each task 🛠️
|
||||
- Activated appropriate experts (security, performance, etc.) 🎭
|
||||
- Activated appropriate specialized agents 🎭
|
||||
- Applied intelligent flags and optimizations ⚡
|
||||
- Provided evidence-based suggestions 📊
|
||||
- Saved insights for future sessions 💾
|
||||
|
||||
**See how easy that was?** No studying required - SuperClaude figures out the complexity so you don't have to.
|
||||
|
||||
@@ -70,69 +74,87 @@ Want to understand how it works? Keep reading. Want to just keep experimenting?
|
||||
|
||||
### What is SuperClaude Really? 🤔
|
||||
|
||||
SuperClaude makes Claude Code smarter for development work. Instead of generic responses, you get specialized help from different experts (security, performance, frontend, etc.) who know their stuff.
|
||||
SuperClaude makes Claude Code smarter for development work. Instead of generic responses, you get specialized help from different agents (security, performance, frontend, etc.) who know their stuff, plus session persistence and behavioral intelligence.
|
||||
|
||||
**The honest truth**: We just released v3.0 and it's fresh out of beta. It works pretty well for what it does, but you should expect some rough edges as we continue improving things. We built this because we wanted Claude Code to be more helpful for real software development workflows.
|
||||
**The honest truth**: V4 Beta represents a major architectural evolution with new session lifecycle management, behavioral modes, and an advanced hooks system. While still in beta, it's significantly more capable than v3, with better context management and intelligent agent coordination.
|
||||
|
||||
**The neat part?** You don't need to manage any of this complexity. Just use normal commands like `/analyze` or `/build` and SuperClaude usually figures out which experts to involve and what tools to use. 🪄
|
||||
|
||||
### What SuperClaude Adds ✨
|
||||
|
||||
**🛠️ 17 Specialized Commands**
|
||||
- Planning tools: `/workflow` (NEW!), `/estimate`, `/task`
|
||||
- Development tools: `/implement`, `/build`, `/design`
|
||||
- Analysis tools: `/analyze`, `/troubleshoot`, `/explain`
|
||||
- Quality tools: `/improve`, `/cleanup`, `/test`
|
||||
- Plus utilities for documentation, git, deployment, and more
|
||||
**🛠️ 21 Specialized Commands**
|
||||
- **Planning tools**: `/estimate`, `/task`, `/brainstorm` (NEW!)
|
||||
- **Development tools**: `/implement`, `/build`, `/design`, `/select-tool` (NEW!)
|
||||
- **Analysis tools**: `/analyze`, `/troubleshoot`, `/explain`
|
||||
- **Quality tools**: `/improve`, `/cleanup`, `/test`
|
||||
- **Session tools**: `/load` (NEW!), `/save` (NEW!), `/reflect` (NEW!)
|
||||
- **Plus utilities** for documentation, git, deployment, and more
|
||||
- **You just use them** - SuperClaude handles the complexity automatically
|
||||
- **NEW**: `/workflow` command for PRD-to-implementation planning
|
||||
- **NEW**: `/implement` command for feature creation (restores v2 functionality)
|
||||
- **V4 NEW**: Session lifecycle commands for persistent context
|
||||
- **V4 NEW**: Interactive brainstorming and reflection capabilities
|
||||
|
||||
**🎭 11 Smart Personas** *(that know when to jump in)*
|
||||
- AI specialists that adapt behavior for different domains
|
||||
- **Auto-activate based on your requests** (security expert for security tasks, etc.)
|
||||
- Manual control available, but usually not needed
|
||||
**🤖 13 Specialized Agents** *(that know when to jump in)*
|
||||
- AI agents that adapt behavior for different domains
|
||||
- **Auto-activate based on your requests** (security agent for security tasks, etc.)
|
||||
- Manual control available, but usually not needed
|
||||
- Think of it as having a whole dev team that knows when to help
|
||||
- **V4 NEW**: Enhanced agent system with better specialization
|
||||
|
||||
**🔧 MCP Server Integration** *(smart external tools)*
|
||||
- Context7: Official library documentation lookup
|
||||
- Sequential: Complex multi-step analysis
|
||||
- Magic: Modern UI component generation
|
||||
- Playwright: Browser automation and testing
|
||||
**🔧 MCP Server Integration** *(6 smart external tools)*
|
||||
- **Context7**: Official library documentation lookup
|
||||
- **Sequential**: Complex multi-step analysis
|
||||
- **Magic**: Modern UI component generation
|
||||
- **Playwright**: Browser automation and testing
|
||||
- **Morphllm**: Intelligent file editing (NEW!)
|
||||
- **Serena**: Semantic code analysis and memory (NEW!)
|
||||
- **Auto-connects when needed** - you don't manage this stuff
|
||||
|
||||
**📋 Enhanced Task Management** *(happens behind the scenes)*
|
||||
- Progress tracking with TodoRead/TodoWrite
|
||||
- Multi-session project management with `/task`
|
||||
- Complex orchestration with `/spawn`
|
||||
- Iterative improvement with `/loop`
|
||||
- **Mostly automatic** - SuperClaude tracks what you're doing
|
||||
**🧠 4 Behavioral Modes** *(intelligent adaptation)*
|
||||
- **Brainstorming Mode**: Interactive requirements discovery (NEW!)
|
||||
- **Introspection Mode**: Meta-cognitive analysis and debugging (NEW!)
|
||||
- **Task Management Mode**: Multi-layer orchestration with wave systems (NEW!)
|
||||
- **Token Efficiency Mode**: Smart compression and optimization (NEW!)
|
||||
- **Auto-activate based on context** - you don't configure them
|
||||
|
||||
**⚡ Token Optimization** *(smart efficiency)*
|
||||
- Smart compression when context gets full
|
||||
- Symbol system for efficient communication
|
||||
- Performance optimization for large operations
|
||||
- **Usually activates** when needed for large projects
|
||||
**🏗️ Session Lifecycle System** *(persistent intelligence)*
|
||||
- Session initialization with `/sc:load` (NEW!)
|
||||
- Persistent context and memory across sessions (NEW!)
|
||||
- Automatic checkpoints and progress tracking (NEW!)
|
||||
- Session reflection and insights with `/sc:reflect` (NEW!)
|
||||
- **Cross-session learning** - SuperClaude remembers and improves
|
||||
|
||||
### Current Status (v3.0) 📊
|
||||
**🔌 Hooks System** *(extensible architecture)*
|
||||
- Performance monitoring and optimization (NEW!)
|
||||
- Quality gates with 8-step validation (NEW!)
|
||||
- Framework coordination and orchestration (NEW!)
|
||||
- Session lifecycle integration (NEW!)
|
||||
- **SuperClaude-Lite** variant for streamlined usage (NEW!)
|
||||
|
||||
### Current Status (V4 Beta) 📊
|
||||
|
||||
**✅ What's Working Well:**
|
||||
- Installation system (completely rewritten, much more reliable)
|
||||
- Core framework with 16 commands and 11 personas
|
||||
- MCP server integration (mostly working)
|
||||
- Basic task management and workflow automation
|
||||
- Documentation and user guides
|
||||
- Complete session lifecycle with persistent memory
|
||||
- 21 commands including new brainstorming and reflection tools
|
||||
- 13 specialized agents with enhanced coordination
|
||||
- 6 MCP servers with Morphllm and Serena integration
|
||||
- 4 behavioral modes with intelligent auto-activation
|
||||
- Advanced hooks system with performance monitoring
|
||||
- SuperClaude-Lite for streamlined workflows
|
||||
|
||||
**⚠️ What's Still Rough:**
|
||||
- This is an initial release - bugs are expected
|
||||
- Some MCP integrations could be smoother
|
||||
- Performance isn't optimized yet for all operations
|
||||
- Some advanced features are experimental
|
||||
**⚠️ What's Still Beta:**
|
||||
- Session persistence optimization (performance tuning ongoing)
|
||||
- Behavioral mode coordination (refining trigger patterns)
|
||||
- Hooks system reliability (extensive testing in progress)
|
||||
- Agent handoff protocols (improving seamless transitions)
|
||||
|
||||
**❌ What We Removed:**
|
||||
- Hooks system (got too complex, coming back in v4)
|
||||
**🚀 Major V4 Improvements:**
|
||||
- **Session Intelligence**: Persistent context across sessions
|
||||
- **Behavioral Modes**: Adaptive intelligence based on task type
|
||||
- **Enhanced Memory**: Cross-session learning and pattern recognition
|
||||
- **Advanced Orchestration**: Multi-layer task coordination
|
||||
- **Extensible Architecture**: Robust hooks system for customization
|
||||
|
||||
We're pretty happy with v3 as a foundation, but there's definitely room for improvement.
|
||||
V4 Beta represents a significant architectural evolution with much more sophisticated context management and intelligent coordination.
|
||||
|
||||
### How It Works 🔄
|
||||
|
||||
@@ -153,12 +175,13 @@ The nice thing is that most of this usually happens automatically. You make a re
|
||||
|
||||
| Component | What It Does | Learn More *(optional!)* |
|
||||
|-----------|--------------|------------|
|
||||
| **Commands** | 15 specialized tools that auto-activate | [Commands Guide](commands-guide.md) |
|
||||
| **Commands** | 21 specialized tools that auto-activate | [Commands Guide](commands-guide.md) |
|
||||
| **Flags** | Modifiers that mostly activate automatically | [Flags Guide](flags-guide.md) |
|
||||
| **Personas** | 11 AI specialists that know when to help | [Personas Guide](personas-guide.md) |
|
||||
| **MCP Servers** | External integrations that connect when useful | [This guide](#core-components-🧩) |
|
||||
| **Modes** | 3 operational modes for different workflows | [This guide](#the-three-operational-modes-🎭) |
|
||||
| **Orchestrator** | The smart routing that makes it all work | [This guide](#the-orchestrator-system-🎯) |
|
||||
| **Agents** | 13 AI specialists that know when to help | [This guide](#agents-ai-specialists-) |
|
||||
| **MCP Servers** | 6 external integrations that connect when useful | [This guide](#mcp-servers-external-capabilities-) |
|
||||
| **Behavioral Modes** | 4 adaptive modes for different workflows | [This guide](#the-four-behavioral-modes-) |
|
||||
| **Session Lifecycle** | Persistent context across sessions | [This guide](#session-lifecycle-system-) |
|
||||
| **Orchestrator** | The smart routing that makes it all work | [This guide](#the-orchestrator-system-) |
|
||||
|
||||
**Remember**: You can use SuperClaude effectively without reading any of these guides. They're here when you get curious about how it works! 🎪
|
||||
|
||||
@@ -172,11 +195,12 @@ SuperClaude is built from several interconnected systems that work together. Her
|
||||
|
||||
Commands are specialized tools that handle specific types of development work. Instead of generic "help me with this," you get purpose-built tools for different scenarios.
|
||||
|
||||
**15 Commands Organized by Purpose:**
|
||||
**21 Commands Organized by Purpose:**
|
||||
|
||||
**Development** 🔨
|
||||
- `/build` - Project building, compilation, bundling
|
||||
- `/design` - System architecture and component design
|
||||
- `/implement` - Feature creation and code generation
|
||||
|
||||
**Analysis** 🔍
|
||||
- `/analyze` - Comprehensive code and system analysis
|
||||
@@ -188,12 +212,20 @@ Commands are specialized tools that handle specific types of development work. I
|
||||
- `/cleanup` - Technical debt reduction
|
||||
- `/test` - Testing and coverage analysis
|
||||
|
||||
**Session Management** 🧠 *(NEW in V4!)*
|
||||
- `/load` - Initialize session with persistent context
|
||||
- `/save` - Save session state and insights
|
||||
- `/reflect` - Task validation and completion analysis
|
||||
|
||||
**Planning & Discovery** 🎯
|
||||
- `/brainstorm` - Interactive requirements discovery *(NEW!)*
|
||||
- `/estimate` - Project estimation and planning
|
||||
- `/task` - Long-term project management
|
||||
|
||||
**Utilities** 🔧
|
||||
- `/document` - Documentation creation
|
||||
- `/git` - Enhanced git workflows
|
||||
- `/load` - Project context loading
|
||||
- `/estimate` - Project estimation
|
||||
- `/task` - Long-term project management
|
||||
- `/select-tool` - Smart tool selection assistance *(NEW!)*
|
||||
- `/spawn` - Complex operation orchestration
|
||||
- `/index` - Command navigation and help
|
||||
|
||||
@@ -232,11 +264,11 @@ Flags change how SuperClaude processes your requests. They're like command-line
|
||||
|
||||
Flags often auto-activate based on context. For example, security-related requests usually get `--persona-security` and `--focus security`. See the [Flags Guide](flags-guide.md) for comprehensive details and patterns.
|
||||
|
||||
### Personas: AI Specialists 🎭
|
||||
### Agents: AI Specialists 🤖
|
||||
|
||||
Personas are like having a team of specialists available on demand. Each brings different expertise, priorities, and approaches to problems.
|
||||
Agents are like having a team of specialists available on demand. Each brings different expertise, priorities, and approaches to problems. V4 Beta features an enhanced agent system with better specialization and coordination.
|
||||
|
||||
**11 Personas Organized by Domain:**
|
||||
**13 Agents Organized by Domain:**
|
||||
|
||||
**Technical Specialists** 🔧
|
||||
- 🏗️ **architect** - Systems design, long-term architecture
|
||||
@@ -255,13 +287,17 @@ Personas are like having a team of specialists available on demand. Each brings
|
||||
- 👨🏫 **mentor** - Education, knowledge transfer
|
||||
- ✍️ **scribe** - Documentation, technical writing
|
||||
|
||||
Personas usually auto-activate based on request patterns but you can override with `--persona-[name]` flags. Each has different priorities (e.g., security persona prioritizes security over speed). See the [Personas Guide](personas-guide.md) for detailed descriptions and examples.
|
||||
**V4 Beta Enhancements** 🆕
|
||||
- 🎯 **brainstorm-PRD** - Requirements analysis and PRD generation *(NEW!)*
|
||||
- 🔮 **workflow-agent** - Advanced workflow orchestration *(NEW!)*
|
||||
|
||||
Agents usually auto-activate based on request patterns but you can override with `--agent-[name]` flags. Each has different priorities (e.g., security agent prioritizes security over speed). V4 Beta improves agent handoff and coordination patterns.
|
||||
|
||||
### MCP Servers: External Capabilities 🔧
|
||||
|
||||
MCP (Model Context Protocol) servers provide specialized capabilities beyond Claude's native abilities.
|
||||
|
||||
**4 Integrated Servers:**
|
||||
**6 Integrated Servers:**
|
||||
|
||||
**Context7** 📚
|
||||
- **Purpose**: Official library documentation and best practices
|
||||
@@ -287,6 +323,18 @@ MCP (Model Context Protocol) servers provide specialized capabilities beyond Cla
|
||||
- **What it provides**: Cross-browser testing, visual validation, metrics
|
||||
- **Example**: `/test e2e --play` runs comprehensive browser tests
|
||||
|
||||
**Morphllm** 🔧 *(NEW in V4!)*
|
||||
- **Purpose**: Intelligent file editing with Fast Apply capability
|
||||
- **When it activates**: Complex refactoring, multi-file edits
|
||||
- **What it provides**: Context-aware code modifications, pattern transformations
|
||||
- **Example**: `/improve legacy-code/ --morph` applies intelligent refactoring
|
||||
|
||||
**Serena** 🧠 *(NEW in V4!)*
|
||||
- **Purpose**: Semantic code analysis and persistent memory
|
||||
- **When it activates**: Project analysis, session persistence, memory operations
|
||||
- **What it provides**: Cross-session context, semantic understanding, project memory
|
||||
- **Example**: `/load --serena` initializes persistent project context
|
||||
|
||||
MCP servers usually coordinate automatically but you can control them with `--all-mcp`, `--no-mcp`, or specific flags like `--c7`.
|
||||
|
||||
### How Components Work Together 🤝
|
||||
@@ -311,17 +359,61 @@ This coordination usually happens for most requests - SuperClaude tries to figur
|
||||
|
||||
---
|
||||
|
||||
## The Three Operational Modes 🎭
|
||||
## The Four Behavioral Modes 🧠
|
||||
|
||||
SuperClaude operates in three distinct modes that optimize different aspects of the development workflow. Understanding these modes helps you get the most out of the framework.
|
||||
SuperClaude V4 Beta features four intelligent behavioral modes that automatically adapt to different types of work. These modes provide specialized behavior patterns while maintaining the same command interface.
|
||||
|
||||
### Task Management Mode 📋
|
||||
### 1. Brainstorming Mode 🎯 *(NEW!)*
|
||||
|
||||
**What it is**: Structured workflow execution with progress tracking and validation.
|
||||
**What it is**: Interactive requirements discovery through collaborative dialogue.
|
||||
|
||||
**When it's used**: Any multi-step operation that needs tracking and coordination.
|
||||
**When it activates**: Vague project requests, exploration keywords, uncertainty indicators.
|
||||
|
||||
**How it works**: SuperClaude breaks work into manageable tasks, tracks progress, and ensures quality through validation gates.
|
||||
**How it works**:
|
||||
- Socratic dialogue to clarify requirements
|
||||
- Interactive exploration of possibilities
|
||||
- Consensus building and brief generation
|
||||
- Automatic handoff to PRD agent
|
||||
|
||||
**Example Usage**:
|
||||
```bash
|
||||
/sc:brainstorm "task management app"
|
||||
# → Interactive dialogue to discover requirements
|
||||
# → Generates comprehensive project brief
|
||||
# → Hands off to brainstorm-PRD agent
|
||||
```
|
||||
|
||||
### 2. Introspection Mode 🧠 *(NEW!)*
|
||||
|
||||
**What it is**: Meta-cognitive analysis of SuperClaude's own reasoning and decision-making.
|
||||
|
||||
**When it activates**: Complex problem-solving, error recovery, framework troubleshooting.
|
||||
|
||||
**How it works**:
|
||||
- Self-reflective analysis of reasoning patterns
|
||||
- Framework compliance validation
|
||||
- Error pattern recognition
|
||||
- Performance optimization insights
|
||||
|
||||
**Example Usage**:
|
||||
```bash
|
||||
/sc:analyze complex-issue/ --introspect
|
||||
# → Meta-cognitive analysis of approach
|
||||
# → Framework compliance checking
|
||||
# → Alternative strategy consideration
|
||||
```
|
||||
|
||||
### 3. Task Management Mode 📋
|
||||
|
||||
**What it is**: Multi-layer orchestration with wave systems and advanced delegation.
|
||||
|
||||
**When it activates**: Multi-step operations, complex projects, large-scale work.
|
||||
|
||||
**How it works**:
|
||||
- Four-layer task hierarchy (Session → Project → Orchestration → Iterative)
|
||||
- Wave orchestration for compound intelligence
|
||||
- Sub-agent delegation for parallel processing
|
||||
- Real-time analytics and performance monitoring
|
||||
|
||||
#### Four Layers of Task Management
|
||||
|
||||
@@ -369,38 +461,196 @@ SuperClaude operates in three distinct modes that optimize different aspects of
|
||||
# → Iteratively improves code with validation between cycles
|
||||
```
|
||||
|
||||
#### Task State Management
|
||||
### 4. Token Efficiency Mode ⚡ *(NEW!)*
|
||||
|
||||
**Core Principles**:
|
||||
- **Evidence-Based Progress**: Measurable outcomes, not just activity
|
||||
- **Single Focus Protocol**: Only one task in_progress at a time
|
||||
- **Real-Time Updates**: Immediate status changes as work progresses
|
||||
- **Quality Gates**: Validation before marking tasks complete
|
||||
**What it is**: Intelligent token optimization with adaptive compression strategies.
|
||||
|
||||
**Task Detection**:
|
||||
- Multi-step operations (3+ steps) → Creates task breakdown
|
||||
- Keywords: build, implement, create, fix, optimize → Activates task tracking
|
||||
- Scope indicators: system, feature, comprehensive → Adds progress monitoring
|
||||
**When it activates**: Context usage >75%, large-scale operations, resource constraints.
|
||||
|
||||
### Introspection Mode 🧠
|
||||
**How it works**:
|
||||
- 5-level compression strategy (Minimal → Emergency)
|
||||
- Symbol systems for efficient communication
|
||||
- Adaptive compression based on content type
|
||||
- Quality-gated validation (≥95% information preservation)
|
||||
|
||||
**What it is**: Meta-cognitive analysis that lets SuperClaude examine its own reasoning and decision-making processes.
|
||||
**Example Features**:
|
||||
```bash
|
||||
# Automatically activates for large operations
|
||||
/sc:analyze enterprise-codebase/
|
||||
# → Token Efficiency Mode activates
|
||||
# → Applies symbol systems and compression
|
||||
# → Maintains quality while reducing tokens by 30-50%
|
||||
```
|
||||
|
||||
**When it's used**: Complex problem-solving, framework troubleshooting, learning moments, or when you explicitly request it with `--introspect`.
|
||||
**Compression Techniques**:
|
||||
- **Smart Symbol Systems**: →, ⇒, ✅, ❌, 🔄 for status and flow
|
||||
- **Intelligent Abbreviations**: cfg, impl, perf, sec for common terms
|
||||
- **Selective Compression**: Framework content preserved, operational data compressed
|
||||
- **Quality Validation**: Real-time monitoring of information preservation
|
||||
|
||||
**How it works**: SuperClaude steps outside normal operation to analyze its thinking patterns, decision logic, and action sequences.
|
||||
---
|
||||
|
||||
#### Core Capabilities
|
||||
## Session Lifecycle System 🏗️ *(NEW in V4!)*
|
||||
|
||||
**Reasoning Analysis** 🧠
|
||||
- Examines logical flow and decision rationale
|
||||
- Evaluates chain of thought coherence
|
||||
- Identifies assumptions and potential biases
|
||||
- Validates reasoning against evidence
|
||||
SuperClaude V4 Beta introduces persistent session management that remembers your work across multiple sessions, enabling continuous learning and context preservation.
|
||||
|
||||
**Action Sequence Review** 🔄
|
||||
- Analyzes tool selection effectiveness
|
||||
- Reviews workflow patterns and efficiency
|
||||
### Core Concept
|
||||
|
||||
```
|
||||
┌─────────────┐ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐
|
||||
│ /sc:load │────▶│ WORK │────▶│ /sc:save │────▶│ NEXT │
|
||||
│ (INIT) │ │ (ACTIVE) │ │ (CHECKPOINT)│ │ SESSION │
|
||||
└─────────────┘ └─────────────┘ └─────────────┘ └─────────────┘
|
||||
│ │
|
||||
└──────────────────── Enhanced Context ───────────────────────┘
|
||||
```
|
||||
|
||||
### Session States
|
||||
|
||||
**1. INITIALIZING** (`/sc:load`)
|
||||
- Activate project via Serena memory system
|
||||
- Load existing memories and context
|
||||
- Initialize session metadata and tracking
|
||||
- **Performance Target**: <500ms
|
||||
|
||||
**2. ACTIVE** (Working Session)
|
||||
- Full project context available
|
||||
- Changes tracked for persistence
|
||||
- Decisions logged for replay
|
||||
- Automatic checkpoint triggers
|
||||
|
||||
**3. CHECKPOINTED** (`/sc:save`)
|
||||
- Analyze session changes and insights
|
||||
- Persist discoveries to memory
|
||||
- Create checkpoint records
|
||||
- Generate summaries if requested
|
||||
|
||||
**4. RESUMED** (Next `/sc:load`)
|
||||
- Load latest checkpoint and context
|
||||
- Restore session state and data
|
||||
- Display resumption summary
|
||||
- Continue from previous state
|
||||
|
||||
### Key Session Commands
|
||||
|
||||
**`/sc:load` - Session Initialization**
|
||||
```bash
|
||||
/sc:load # Initialize current project
|
||||
/sc:load --deep # Deep context loading
|
||||
/sc:load --resume # Resume from last checkpoint
|
||||
```
|
||||
|
||||
**`/sc:save` - Session Persistence**
|
||||
```bash
|
||||
/sc:save # Save current session state
|
||||
/sc:save --checkpoint # Create checkpoint
|
||||
/sc:save --summarize # Include session summary
|
||||
```
|
||||
|
||||
**`/sc:reflect` - Session Analysis**
|
||||
```bash
|
||||
/sc:reflect --type task # Validate current task
|
||||
/sc:reflect --type session # Analyze session progress
|
||||
/sc:reflect --type completion # Check if work is complete
|
||||
```
|
||||
|
||||
### Automatic Checkpoints
|
||||
|
||||
SuperClaude automatically creates checkpoints:
|
||||
- **Task-Based**: Major task completion
|
||||
- **Time-Based**: Every 30 minutes of active work
|
||||
- **Risk-Based**: Before high-risk operations
|
||||
- **Error Recovery**: After recovering from errors
|
||||
|
||||
### Session Memory Organization
|
||||
|
||||
```
|
||||
memories/
|
||||
├── session/
|
||||
│ ├── current # Latest session
|
||||
│ ├── {timestamp} # Individual sessions
|
||||
│ └── history/ # Archived sessions
|
||||
├── checkpoints/
|
||||
│ ├── latest # Latest checkpoint
|
||||
│ ├── {timestamp} # Individual checkpoints
|
||||
│ └── task-{id} # Task-specific checkpoints
|
||||
├── summaries/
|
||||
│ ├── daily/{date} # Daily summaries
|
||||
│ └── insights/{topic} # Topical insights
|
||||
└── project_state/
|
||||
├── context_enhanced # Accumulated context
|
||||
└── decisions_log # Architecture decisions
|
||||
```
|
||||
|
||||
### Benefits
|
||||
|
||||
**Cross-Session Learning**
|
||||
- SuperClaude remembers project patterns and decisions
|
||||
- Context builds over time rather than starting fresh
|
||||
- Previous work informs future recommendations
|
||||
|
||||
**Reliable Progress Tracking**
|
||||
- Work is never lost due to session interruptions
|
||||
- Resume exactly where you left off
|
||||
- Complete audit trail of decisions and changes
|
||||
|
||||
**Enhanced Context Understanding**
|
||||
- Deeper project comprehension over time
|
||||
- Pattern recognition across multiple sessions
|
||||
- Better recommendations based on project history
|
||||
|
||||
---
|
||||
|
||||
## Hooks System & SuperClaude-Lite 🔌 *(NEW in V4!)*
|
||||
|
||||
V4 Beta introduces an advanced hooks system for extensibility and SuperClaude-Lite for streamlined workflows.
|
||||
|
||||
### Hooks System Architecture
|
||||
|
||||
**8-Step Quality Gates**
|
||||
1. **Syntax Validation**: Language-specific syntax checking
|
||||
2. **Type Analysis**: Type compatibility and inference
|
||||
3. **Code Quality**: Linting rules and quality standards
|
||||
4. **Security Assessment**: Vulnerability analysis and compliance
|
||||
5. **E2E Testing**: End-to-end test execution and validation
|
||||
6. **Performance Analysis**: Performance benchmarking and optimization
|
||||
7. **Documentation**: Documentation completeness and accuracy
|
||||
8. **Integration Testing**: Deployment and cross-browser validation
|
||||
|
||||
**Performance Monitoring**
|
||||
- Real-time operation timing and resource usage
|
||||
- Automatic performance optimization suggestions
|
||||
- Bottleneck detection and resolution
|
||||
- Quality preservation metrics
|
||||
|
||||
**Framework Coordination**
|
||||
- Seamless integration with all SuperClaude components
|
||||
- Session lifecycle management and checkpointing
|
||||
- Cross-mode behavioral coordination
|
||||
- MCP server orchestration
|
||||
|
||||
### SuperClaude-Lite Variant
|
||||
|
||||
**What it is**: Streamlined version of SuperClaude optimized for specific workflows.
|
||||
|
||||
**Key Features**:
|
||||
- **Reduced complexity**: Simplified command set focused on core functionality
|
||||
- **Faster performance**: Optimized for speed with minimal overhead
|
||||
- **Targeted use cases**: Specialized for quick analysis, code review, and documentation
|
||||
- **Easy deployment**: Lighter installation with fewer dependencies
|
||||
|
||||
**When to use SuperClaude-Lite**:
|
||||
- Quick code reviews and analysis
|
||||
- Documentation generation workflows
|
||||
- Performance-critical environments
|
||||
- Teams wanting focused functionality
|
||||
- CI/CD pipeline integration
|
||||
|
||||
**Relationship to Full SuperClaude**:
|
||||
- Shares core architecture and quality standards
|
||||
- Compatible session formats and memory systems
|
||||
- Can upgrade to full SuperClaude when needed
|
||||
- Maintains behavioral consistency
|
||||
- Considers alternative approaches
|
||||
- Identifies optimization opportunities
|
||||
|
||||
@@ -956,21 +1206,42 @@ If you haven't installed SuperClaude yet, see the [Installation Guide](installat
|
||||
```bash
|
||||
# Test basic functionality
|
||||
/sc:help # Should show SuperClaude commands
|
||||
/sc:load # Initialize session (NEW!)
|
||||
/sc:analyze README.md # Try analyzing a simple file
|
||||
/sc:build --help # Check command options
|
||||
/sc:save # Save session state (NEW!)
|
||||
```
|
||||
|
||||
#### V4 Beta Session Workflow
|
||||
Experience the new session persistence:
|
||||
|
||||
```bash
|
||||
# Initialize your first session
|
||||
/sc:load --deep # Deep project context loading
|
||||
|
||||
# Try the new brainstorming command
|
||||
/sc:brainstorm "improve this project" # Interactive requirements discovery
|
||||
|
||||
# Work with enhanced context
|
||||
/sc:analyze codebase/ --focus architecture # Persistent context helps
|
||||
|
||||
# Save your session
|
||||
/sc:save --checkpoint # Persistent memory for next session
|
||||
```
|
||||
|
||||
#### Understanding Auto-Activation
|
||||
Try these commands to see how SuperClaude automatically chooses the right tools:
|
||||
|
||||
```bash
|
||||
# Frontend work → frontend persona + Magic MCP
|
||||
# Frontend work → frontend agent + Magic MCP
|
||||
/sc:build src/components/
|
||||
|
||||
# Security analysis → security persona + Sequential MCP
|
||||
# Security analysis → security agent + Sequential MCP
|
||||
/sc:analyze auth/ --focus security
|
||||
|
||||
# Performance investigation → performance persona + Playwright MCP
|
||||
# Complex refactoring → Morphllm MCP + intelligent editing
|
||||
/sc:improve legacy-code/ --focus quality
|
||||
|
||||
# Performance investigation → performance agent + Playwright MCP
|
||||
/sc:analyze --focus performance slow-endpoints/
|
||||
```
|
||||
|
||||
@@ -978,46 +1249,59 @@ Watch for auto-activated flags and personas in the output. This shows SuperClaud
|
||||
|
||||
### Development Workflow Patterns 🔄
|
||||
|
||||
#### New Project Onboarding
|
||||
#### New Project Onboarding *(Enhanced in V4!)*
|
||||
When starting work on an unfamiliar project:
|
||||
|
||||
```bash
|
||||
# 1. Load project context
|
||||
# 1. Initialize persistent session
|
||||
/sc:load --deep --summary
|
||||
# → Serena MCP loads project context persistently
|
||||
# → Gives overview of structure, dependencies, patterns
|
||||
|
||||
# 2. Analyze architecture
|
||||
# 2. Interactive project discovery
|
||||
/sc:brainstorm "understand this project"
|
||||
# → Brainstorming Mode helps discover project goals
|
||||
# → Interactive dialogue to understand requirements
|
||||
|
||||
# 3. Analyze architecture with enhanced context
|
||||
/sc:analyze --focus architecture
|
||||
# → 🏗️ architect persona provides system understanding
|
||||
# → 🏗️ architect agent with persistent memory
|
||||
# → Serena provides semantic understanding
|
||||
|
||||
# 3. Check code quality
|
||||
/sc:analyze --focus quality
|
||||
# → 🧪 qa persona identifies potential issues
|
||||
|
||||
# 4. Review documentation
|
||||
/sc:document README --type guide
|
||||
# → ✍️ scribe persona improves project documentation
|
||||
# 4. Save discoveries for future sessions
|
||||
/sc:save --checkpoint --summarize
|
||||
# → Persistent memory of project insights
|
||||
# → Available in future sessions
|
||||
```
|
||||
|
||||
#### Feature Development Cycle
|
||||
For developing new features:
|
||||
|
||||
```bash
|
||||
# 1. Design phase
|
||||
# 1. Requirements discovery (NEW!)
|
||||
/sc:brainstorm "user dashboard feature"
|
||||
# → Interactive requirements gathering
|
||||
# → Generates comprehensive feature brief
|
||||
|
||||
# 2. Design with persistent context
|
||||
/sc:design user-dashboard --type component
|
||||
# → 🏗️ architect + 🎨 frontend personas coordinate
|
||||
# → 🏗️ architect + 🎨 frontend agents coordinate
|
||||
# → Leverages session memory for consistency
|
||||
|
||||
# 2. Implementation
|
||||
/sc:build dashboard-components/
|
||||
# → 🎨 frontend persona + Magic MCP for UI generation
|
||||
# 3. Implementation with intelligent editing
|
||||
/sc:build dashboard-components/
|
||||
# → 🎨 frontend agent + Magic MCP for UI generation
|
||||
# → Morphllm for intelligent code modifications
|
||||
|
||||
# 3. Testing
|
||||
# 4. Testing with cross-session context
|
||||
/sc:test --type e2e dashboard/
|
||||
# → 🧪 qa persona + Playwright MCP for testing
|
||||
# → 🧪 qa agent + Playwright MCP for testing
|
||||
# → Session memory ensures test consistency
|
||||
|
||||
# 4. Documentation
|
||||
/sc:document dashboard/ --type api
|
||||
# → ✍️ scribe persona creates comprehensive docs
|
||||
# 5. Reflection and checkpoint
|
||||
/sc:reflect --type completion
|
||||
# → Validates feature completeness
|
||||
# → Creates checkpoint for future work
|
||||
```
|
||||
|
||||
#### Bug Investigation & Resolution
|
||||
@@ -2916,38 +3200,43 @@ We believe SuperClaude can become significantly more helpful for software develo
|
||||
|
||||
## Conclusion 🎉
|
||||
|
||||
You've now got a comprehensive understanding of SuperClaude v3.0 - its components, capabilities, and how to use them effectively. Let's wrap up with the key takeaways that will help you get the most out of the framework.
|
||||
You've now got a comprehensive understanding of SuperClaude V4 Beta - its components, capabilities, and how to use them effectively. Let's wrap up with the key takeaways that will help you get the most out of the framework.
|
||||
|
||||
### Key Takeaways 🎯
|
||||
|
||||
#### SuperClaude's Core Value
|
||||
SuperClaude transforms Claude Code from a general-purpose AI assistant into a specialized development partner through:
|
||||
- **15 specialized commands** that understand development workflows
|
||||
- **11 expert personas** that bring domain-specific knowledge
|
||||
- **Intelligent orchestration** that coordinates tools automatically
|
||||
- **Quality-first approach** that maintains safety and reliability
|
||||
SuperClaude V4 Beta transforms Claude Code into an intelligent development partner through:
|
||||
- **21 specialized commands** including session management and brainstorming
|
||||
- **13 expert agents** with enhanced coordination and specialization
|
||||
- **4 behavioral modes** that adapt intelligently to different work types
|
||||
- **Session persistence** that remembers and learns across sessions
|
||||
- **Advanced orchestration** with hooks system and quality gates
|
||||
- **6 MCP servers** including intelligent editing and semantic analysis
|
||||
|
||||
#### The Power is in the Coordination
|
||||
SuperClaude's power comes not from any single feature, but from how components work together:
|
||||
- Commands usually activate appropriate personas and MCP servers
|
||||
- Personas coordinate with each other for multi-domain problems
|
||||
- The orchestrator optimizes tool selection and resource usage
|
||||
- Quality gates ensure consistent, reliable outcomes
|
||||
SuperClaude V4's power comes from intelligent system integration:
|
||||
- **Session lifecycle** maintains context across all interactions
|
||||
- **Behavioral modes** adapt automatically to different work patterns
|
||||
- **Agents coordinate** seamlessly for multi-domain problems
|
||||
- **MCP servers** integrate intelligently based on task requirements
|
||||
- **Quality gates** ensure consistent, reliable outcomes through 8-step validation
|
||||
- **Memory system** enables continuous learning and improvement
|
||||
|
||||
#### Start Simple, Scale Intelligently
|
||||
The best approach to SuperClaude is progressive:
|
||||
1. **Begin with basic commands** to understand core functionality
|
||||
2. **Trust auto-activation** to learn optimal tool combinations
|
||||
3. **Add manual control** when you need specific perspectives
|
||||
4. **Experiment with advanced features** as your confidence grows
|
||||
The best approach to SuperClaude V4 Beta is progressive:
|
||||
1. **Initialize sessions** with `/sc:load` to experience persistent context
|
||||
2. **Try brainstorming** with `/sc:brainstorm` for interactive discovery
|
||||
3. **Trust behavioral modes** to adapt automatically to your work patterns
|
||||
4. **Use session persistence** with `/sc:save` to build continuous context
|
||||
5. **Experiment with advanced features** like hooks and multi-layer orchestration
|
||||
|
||||
### What Makes SuperClaude Different 🌟
|
||||
|
||||
#### Honest About Limitations
|
||||
- We acknowledge v3.0 is fresh out of beta with rough edges
|
||||
- We clearly document what works well vs. what's still experimental
|
||||
- We prioritize reliability over flashy features
|
||||
- We provide realistic timelines and expectations
|
||||
- V4 Beta represents major architectural improvements but is still in beta
|
||||
- We clearly document what's working well vs. what's still being refined
|
||||
- Session persistence and behavioral modes are sophisticated but still evolving
|
||||
- We prioritize reliability and user experience over flashy features
|
||||
|
||||
#### Evidence-Based Development
|
||||
- All recommendations backed by verifiable data
|
||||
@@ -2965,24 +3254,27 @@ The best approach to SuperClaude is progressive:
|
||||
|
||||
#### For New Users
|
||||
1. **Start with installation**: Follow the [Installation Guide](installation-guide.md)
|
||||
2. **Try basic commands**: `/help`, `/analyze README.md`, `/build --help`
|
||||
3. **Explore domain guides**: [Commands](commands-guide.md), [Flags](flags-guide.md), [Personas](personas-guide.md)
|
||||
4. **Build confidence gradually**: Simple tasks → complex workflows → advanced features
|
||||
2. **Initialize your first session**: `/sc:load` to experience persistent context
|
||||
3. **Try interactive discovery**: `/sc:brainstorm "project idea"` for requirements
|
||||
4. **Experience the lifecycle**: Use `/sc:save` to persist your session
|
||||
5. **Build confidence gradually**: Session basics → behavioral modes → advanced orchestration
|
||||
|
||||
#### For Experienced Users
|
||||
1. **Optimize your workflows**: Identify flag combinations that work well for your needs
|
||||
2. **Experiment with coordination**: Try different persona combinations on complex problems
|
||||
3. **Contribute feedback**: Share what works (and what doesn't) in your environment
|
||||
4. **Explore advanced features**: Wave orchestration, sub-agent delegation, introspection mode
|
||||
1. **Leverage session persistence**: Build long-term project context across sessions
|
||||
2. **Master behavioral modes**: Understand how brainstorming, introspection, task management, and token efficiency work
|
||||
3. **Explore advanced orchestration**: Wave systems, hooks integration, and multi-layer coordination
|
||||
4. **Contribute feedback**: Share what works (and what doesn't) with V4 Beta features
|
||||
5. **Try SuperClaude-Lite**: Experiment with the streamlined variant for specific workflows
|
||||
|
||||
### When to Use SuperClaude 🤔
|
||||
|
||||
#### SuperClaude Excels At
|
||||
- **Development workflows**: Building, testing, deploying, documenting
|
||||
- **Code analysis**: Quality assessment, security scanning, performance optimization
|
||||
- **Learning and understanding**: Explaining complex systems, onboarding to new projects
|
||||
- **Quality improvement**: Systematic refactoring, technical debt reduction
|
||||
- **Multi-domain problems**: Issues requiring multiple types of expertise
|
||||
#### SuperClaude V4 Beta Excels At
|
||||
- **Persistent development workflows**: Long-term projects with cross-session context
|
||||
- **Interactive requirements discovery**: Brainstorming and clarifying project goals
|
||||
- **Intelligent code analysis**: Semantic understanding with Serena and Morphllm
|
||||
- **Adaptive task management**: Multi-layer orchestration with behavioral intelligence
|
||||
- **Cross-domain coordination**: Complex problems requiring multiple agents and modes
|
||||
- **Session-based learning**: Building project understanding over time
|
||||
|
||||
#### When to Use Standard Claude Code
|
||||
- **Simple questions**: Quick explanations that don't need specialized tools
|
||||
@@ -3028,7 +3320,7 @@ The framework gets better through:
|
||||
|
||||
### Final Thoughts 🎉
|
||||
|
||||
SuperClaude v3.0 represents a solid foundation for enhanced software development workflows. While it's not perfect and still has room for improvement, it demonstrates how AI can be thoughtfully integrated into development practices without disrupting existing workflows or replacing human expertise.
|
||||
SuperClaude V4 Beta represents a significant architectural evolution in AI-assisted development. With persistent sessions, behavioral intelligence, and advanced orchestration, it demonstrates how AI can become a true development partner that learns and adapts while respecting existing workflows and human expertise.
|
||||
|
||||
The framework succeeds when it makes you more productive, helps you learn new things, or catches issues you might have missed. It's designed to be a helpful colleague rather than a replacement for understanding your craft.
|
||||
|
||||
@@ -3042,7 +3334,7 @@ Whether you use SuperClaude occasionally for specific tasks or integrate it deep
|
||||
|
||||
---
|
||||
|
||||
*Last updated: July 2024*
|
||||
*SuperClaude v3.0 User Guide*
|
||||
*Last updated: January 2025*
|
||||
*SuperClaude V4 Beta User Guide*
|
||||
|
||||
*For questions, feedback, or contributions, visit our GitHub repository or join the community discussions. We're always happy to hear from users and learn about your experiences with the framework.*
|
||||
|
||||
Reference in New Issue
Block a user