Skip to content

Latest commit

 

History

History
241 lines (204 loc) · 7.99 KB

WORKFLOW_COMMANDS.md

File metadata and controls

241 lines (204 loc) · 7.99 KB

WORKFLOW COMMANDS - Slash Command Orchestration

Mission Statement

IMPERATIVE: Orchestrate complex workflows through slash command composition. Multi-step patterns, decision trees, and workflow automation using Claude Code command system.

Core Workflow Philosophy

  • Compose Commands: Chain slash commands for complex multi-step workflows
  • Decision-Driven: Use conditional logic for workflow branching
  • Pattern-Based: Reusable workflow templates for systematic execution
  • Reference-Based: Individual commands in .claude/commands/ contain detailed execution

Slash Command Quick Reference

Session Management

  • /session-start - Initialize session with tools and context
  • /session-end - Close session with memory sync and staging
  • /context-switch - Switch between projects or contexts
  • /emergency-restore - Recover from session/context failures

Development Workflows

  • /analyze-codebase - Systematic code analysis and architecture understanding
  • /implement-feature - 4-phase feature development with quality gates
  • /debug-issue - Systematic debugging with root cause analysis
  • /review-code - Self-critique review process with improvement cycle
  • /refactor-code - Safe refactoring with regression prevention
  • /explore-code - Code exploration for mental model building
  • /modify-code - Symbol-based code modification with validation

Research & Analysis

  • /research-library-docs - Context7 documentation research for libraries
  • /research-web - Playwright web research for live sites and demos
  • /analyze-requirements - Requirements breakdown with decision framework
  • /investigate-performance - Performance analysis and optimization

Quality Assurance

  • /assess-quality - 4-dimension quality assessment with scoring
  • /apply-critique - Self-critique cycle for systematic improvement
  • /validate-completion - Completion verification with readiness check

Memory Management

  • /capture-insight - High-entropy insight capture with cross-referencing
  • /recall-context - Context restoration from distributed memory layers
  • /sync-consciousness - Memory synchronization across all layers
  • /clean-memory - Memory maintenance and consolidation

Tool Execution

  • /test-browser - Browser testing with systematic interaction sequences
  • /modify-code - Code modification through symbol-based editing

Error Recovery

  • /recover-tool-failure - Tool failure recovery with fallback procedures
  • /handle-context-overload - Context optimization for information overload
  • /resolve-memory-inconsistency - Memory reconciliation across layers

Multi-Command Workflow Patterns

Complete Feature Development Flow

1. /session-start
2. /analyze-codebase [target_area]
3. /research-library-docs [dependencies] (if needed)
4. /implement-feature [feature_name]
5. /review-code [implementation]
6. /test-browser [feature_testing] (if UI involved)
7. /validate-completion [feature]
8. /session-end

Debugging & Investigation Flow

1. /debug-issue [problem_description]
2. /investigate-performance [area] (if performance-related)
3. /modify-code [fix_implementation]
4. /test-browser [validation] (if UI involved)
5. /apply-critique [fix_review]
6. /validate-completion [fix]

Research & Implementation Flow

1. /research-library-docs [library_name]
2. /research-web [demo_sites] (if needed)
3. /analyze-requirements [integration_specs]
4. /implement-feature [integration]
5. /assess-quality [implementation]

Quality Improvement Flow

1. /assess-quality [target_component]
2. /apply-critique [identified_issues]
3. /refactor-code [improvements] (if needed)
4. /validate-completion [improved_version]

Memory Management Flow

1. /recall-context [relevant_domain]
2. [Execute main work]
3. /capture-insight [valuable_discoveries]
4. /sync-consciousness
5. /clean-memory [outdated_items] (periodic)

Conditional Workflow Decision Trees

Research Decision Tree

Need information about:
├── Library/API documentation? → /research-library-docs
├── Live website/demo functionality? → /research-web  
├── Performance characteristics? → /investigate-performance
└── Requirements analysis? → /analyze-requirements

Development Path Selection

Starting development work:
├── New feature? → /implement-feature
├── Bug fix needed? → /debug-issue → /modify-code
├── Code quality issues? → /apply-critique → /refactor-code
└── Exploration needed? → /analyze-codebase → /explore-code

Quality Gate Decision Tree

Quality assessment results:
├── Score ≥ 7.5? → /validate-completion
├── Score 6.0-7.4? → /apply-critique → re-assess
├── Score < 6.0? → /refactor-code → /apply-critique → re-assess
└── Critical issues? → /debug-issue → restart workflow

Error Recovery Decision Tree

Encountering errors:
├── Tool failure? → /recover-tool-failure
├── Context overload? → /handle-context-overload  
├── Memory inconsistency? → /resolve-memory-inconsistency
└── Session corruption? → /emergency-restore

Advanced Workflow Automation

Parallel Command Execution

Concurrent research workflow:
- Terminal 1: /research-library-docs [primary_library]
- Terminal 2: /research-web [competitor_analysis]  
- Terminal 3: /analyze-codebase [integration_points]
→ Synthesize results → /implement-feature

Iterative Improvement Cycle

Quality improvement loop:
1. /assess-quality [component] 
2. If score < 7.5: /apply-critique [issues]
3. /modify-code [improvements]
4. Repeat step 1 until score ≥ 7.5
5. /validate-completion [final_version]

Context-Aware Workflow Selection

Based on ACTIVE_CONTEXT.md objectives:
├── Research phase? → Research Flow pattern
├── Implementation phase? → Development Flow pattern  
├── Quality phase? → Quality Improvement Flow pattern
└── Completion phase? → Validation and documentation

Workflow Efficiency Patterns

Batched Operations

Memory management session:
/recall-context [project] → 
/capture-insight [discovery1] → 
/capture-insight [discovery2] → 
/sync-consciousness → 
/clean-memory [obsolete]

Tool-Specific Workflows

Browser-heavy workflow:
/research-web [target_site] → 
/test-browser [functionality] → 
/implement-feature [ui_components] → 
/test-browser [validation]

Research-heavy workflow:  
/research-library-docs [lib1] →
/research-library-docs [lib2] →
/analyze-requirements [integration] →
/implement-feature [solution]

Recovery-Enabled Workflows

Resilient development pattern:
1. /session-start
2. Try: /implement-feature [name]
3. On failure: /recover-tool-failure → retry
4. On quality issues: /apply-critique → improve
5. On completion: /validate-completion
6. /session-end

Workflow Customization Guidelines

Creating Custom Workflow Patterns

  1. Identify Repetitive Sequences: Look for command combinations used frequently
  2. Define Decision Points: Add conditional logic for workflow branching
  3. Include Error Handling: Plan recovery commands for common failure modes
  4. Optimize for Context: Consider token efficiency and tool switching overhead
  5. Document Pattern: Create reusable template with parameter placeholders

Workflow Pattern Template

### [Pattern Name] Flow
**Purpose**: [Brief description]
**Triggers**: [When to use this workflow]
**Prerequisites**: [Required context or setup]

Workflow:
1. /command1 [parameter]
2. /command2 [parameter] 
3. [Conditional logic if needed]
4. /command3 [parameter]

**Success Criteria**: [How to know workflow completed successfully]
**Recovery**: [What to do if workflow fails]

Workflow Philosophy: Compose commands systematically, handle errors gracefully, optimize for efficiency Command Integration: Individual commands provide execution, workflows provide orchestration Pattern Evolution: Document successful patterns, iterate on workflow effectiveness