SuperCharged-Claude-Code-Upgrade/cognitive-core/QUICK-REFERENCE.md

🧠 Cognitive Enhancement Suite - Quick Reference

One-page guide for everyday use

---

🎯 What These Skills Do

Skill	Purpose	When It Activates
cognitive-planner	Analyzes tasks, selects approach	Complex requests, "how should I..."
cognitive-safety	Blocks security vulnerabilities	Writing code, running commands
cognitive-context	Adapts to your language/expertise	All interactions

---

🚀 Quick Start

Just use Claude Code normally - skills activate automatically.

You: "Add user authentication to my app"
    ↓
Cognitive skills analyze + protect + adapt
    ↓
Superpowers executes with TDD
    ↓
Secure, tested code

---

💬 Example Commands

For Planning

"How should I build a realtime chat system?"
"Break this down: Add payment processing"
"What's the best approach for file uploads?"

For Safety

"Review this code for security issues"
"Is this command safe to run?"
"Check for vulnerabilities in this function"

For Context

"Explain React hooks like I'm a beginner"
"Give me the expert-level explanation"
"Explícame cómo funciona Docker en español"

---

🎨 Complexity Levels

Level	Description	Example
Simple	Single file, <50 lines	Add a button
Moderate	2-5 files, 50-200 lines	Add authentication
Complex	5+ files, 200+ lines	Build REST API
Very Complex	Architecture changes	Microservices migration

---

🛡️ Safety Checks (Automatic)

✅ Blocks hardcoded secrets ✅ Prevents SQL injection ✅ Prevents XSS vulnerabilities ✅ Validates commands before running ✅ Checks dependency security ✅ Enforces best practices

---

🌐 Supported Languages

English, Spanish, French, German, Italian, Portuguese, Chinese, Japanese, Korean, Russian, Arabic, Hindi

Auto-detected from your messages.

---

👥 Expertise Levels

Level	Indicators	Response Style
Beginner	"How do I...", basic questions	Detailed, educational, examples
Intermediate	"Best practice...", "Why..."	Balanced, explains reasoning
Expert	"Optimize...", specific technical	Concise, advanced topics

Auto-detected and adapted to.

---

📋 Workflow Integration

YOUR REQUEST
     ↓
┌─────────────────┐
│ COGNITIVE-PLANNER │ ← Analyzes complexity
└────────┬────────┘
         ↓
    ┌─────────┐
    │ SUPER-  │ ← Systematic execution
    │ POWERS  │   (if complex)
    └────┬────┘
         ↓
┌─────────────────┐
│ COGNITIVE-SAFETY │ ← Validates security
└────────┬────────┘
         ↓
┌──────────────────┐
│ COGNITIVE-CONTEXT │ ← Adapts to you
└────────┬─────────┘
         ↓
   YOUR RESULT

---

⚡ Pro Tips

1. Be specific → Better planning
2. Ask "why" → Deeper understanding
3. Say your level → Better adaptation
4. Use your language → Auto-detected
5. Trust warnings → Security matters

---

🔧 Customization

# ~/.env
COGNITIVE_PLANNER_THRESHOLD=high    # Only plan complex tasks
COGNITIVE_SAFETY_STRICT_MODE=true   # Block everything risky
COGNITIVE_CONTEXT_LANGUAGE=spanish  # Force language

---

🐛 Common Issues

Issue	Solution
Skills not activating	Check ~/.claude/skills/cognitive-*/ exists
Wrong language	Specify: "Explain in Spanish: ..."
Too much detail	Say: "Give me expert-level explanation"
Too little detail	Say: "Explain like I'm a beginner"
Safety blocking	Say: "I understand this is dev only"

---

📚 Full Documentation

• README.md - Complete guide
• INTEGRATION.md - Workflows and examples
• SKILL.md (each skill) - Detailed behavior

---

🎯 Mental Model

Think of these skills as:

cognitive-planner = Your technical lead

• Plans the approach
• Selects the right tools
• Coordinates execution

cognitive-safety = Your security reviewer

• Checks every line of code
• Blocks vulnerabilities
• Enforces best practices

cognitive-context = Your personal translator

• Understands your level
• Speaks your language
• Adapts explanations

---

✅ Success Indicators

You'll know it's working when:

✅ Tasks are broken down automatically ✅ Security warnings appear before issues ✅ Explanations match your expertise ✅ Your preferred language is used ✅ Superpowers activates for complex tasks ✅ Commands are validated before running

---

🚦 Quick Decision Tree

Need to code?
├─ Simple? → Just do it (with safety checks)
└─ Complex? → Plan → Execute with TDD

Need to debug?
└─ Always → Use systematic debugging

Need to learn?
└─ Always → Adapted to your level

Writing code?
└─ Always → Safety validation

Running commands?
└─ Always → Command safety check

---

💪 Key Benefits

🎯 Autonomous - Works automatically, no commands needed 🛡️ Secure - Prevents vulnerabilities before they happen 🌐 Adaptive - Learns and adapts to you ⚡ Fast - Minimal overhead (~0.4s) 🔗 Integrated - Works with Superpowers seamlessly

---

Just use Claude Code normally - the skills handle the rest! 🧠