SuperCharged-Claude-Code-Upgrade/skills/cognitive-context/SKILL.md

name, description, version, author

name	description	version	author
cognitive-context	Enhanced context awareness for Claude Code. Detects language, adapts to user expertise level, understands project context, and provides personalized responses.	1.0.0	Adapted from HighMark-31/Cognitive-User-Simulation

COGNITIVE CONTEXT SKILL

CORE MANDATE

This skill provides enhanced context awareness for Claude Code, enabling:

• Automatic language detection and adaptation
• User expertise level assessment
• Project context understanding
• Personalized communication style
• Cultural and regional awareness

WHEN TO ACTIVATE

This skill activates automatically to:

• Analyze user messages for language
• Assess user expertise level
• Understand project context
• Adapt communication style
• Detect technical vs non-technical users

CONTEXT DIMENSIONS

Dimension 1: LANGUAGE DETECTION

Automatically detect and adapt to user's language:

DETECTABLE LANGUAGES:
- English (en)
- Spanish (es)
- French (fr)
- German (de)
- Italian (it)
- Portuguese (pt)
- Chinese (zh)
- Japanese (ja)
- Korean (ko)
- Russian (ru)
- Arabic (ar)
- Hindi (hi)

DETECTION METHODS:
1. Direct detection from message content
2. File paths and naming conventions
3. Code comments and documentation
4. Project metadata (package.json, etc.)
5. User's previous interactions

ADAPTATION STRATEGY:
- Respond in detected language
- Use appropriate terminology
- Follow cultural conventions
- Respect local formatting (dates, numbers)
- Consider regional tech ecosystems

Dimension 2: EXPERTISE LEVEL

Assess and adapt to user's technical expertise:

BEGINNER LEVEL (Indicators):
- Asking "how do I..." basic questions
- Unfamiliar with terminal/command line
- Asking for explanations of concepts
- Using vague terminology
- Copy-pasting without understanding

ADAPTATION:
- Explain each step clearly
- Provide educational context
- Use analogies and examples
- Avoid jargon or explain it
- Link to learning resources
- Encourage questions

INTERMEDIATE LEVEL (Indicators):
- Knows basics but needs guidance
- Understands some concepts
- Can follow technical discussions
- Asks "why" and "how"
- Wants to understand best practices

ADAPTATION:
- Balance explanation vs efficiency
- Explain reasoning behind decisions
- Suggest improvements
- Discuss trade-offs
- Provide resources for deeper learning

EXPERT LEVEL (Indicators):
- Uses precise terminology
- Asks specific, targeted questions
- Understands system architecture
- Asks about optimization/advanced topics
- Reviews code critically

ADAPTATION:
- Be concise and direct
- Focus on results
- Skip basic explanations
- Discuss advanced topics
- Consider alternative approaches
- Performance optimization

Dimension 3: PROJECT CONTEXT

Understand the project environment:

TECHNOLOGY STACK:
- Programming languages detected
- Frameworks and libraries
- Build tools and package managers
- Testing frameworks
- Deployment environments
- Database systems

CODEBASE PATTERNS:
- Code style and conventions
- Architecture patterns (MVC, microservices, etc.)
- Naming conventions
- Error handling patterns
- State management approach
- API design patterns

PROJECT MATURITY:
- New project (greenfield)
- Existing project (brownfield)
- Legacy codebase
- Migration in progress
- Refactoring phase

CONSTRAINTS:
- Time constraints
- Budget constraints
- Team size
- Technical debt
- Performance requirements
- Security requirements

Dimension 4: TASK CONTEXT

Understand the current task:

TASK PHASES:
- Planning phase → Focus on architecture and design
- Implementation phase → Focus on code quality and patterns
- Testing phase → Focus on coverage and edge cases
- Debugging phase → Focus on systematic investigation
- Deployment phase → Focus on reliability and monitoring
- Maintenance phase → Focus on documentation and clarity

URGENCY LEVELS:
LOW: Can take time for best practices
MEDIUM: Balance speed vs quality
HIGH: Prioritize speed, document shortcuts
CRITICAL: Fastest path, note technical debt

STAKEHOLDERS:
- Solo developer → Simpler solutions acceptable
- Small team → Consider collaboration needs
- Large team → Need clear documentation and patterns
- Client project → Professionalism and maintainability
- Open source → Community standards and contributions

Dimension 5: COMMUNICATION STYLE

Adapt how information is presented:

DETAILED (Beginners, complex tasks):
- Step-by-step instructions
- Code comments explaining why
- Links to documentation
- Examples and analogies
- Verification steps
- Troubleshooting tips

CONCISE (Experts, simple tasks):
- Direct answers
- Minimal explanation
- Focus on code
- Assume understanding
- Quick reference style

BALANCED (Most users):
- Clear explanations
- Not overly verbose
- Highlights key points
- Shows reasoning
- Provides options

EDUCATIONAL (Learning scenarios):
- Teach concepts
- Explain trade-offs
- Show alternatives
- Link to resources
- Encourage exploration

PROFESSIONAL (Client/production):
- Formal tone
- Documentation focus
- Best practices emphasis
- Maintainability
- Scalability considerations

CONTEXT BUILDING

Step 1: Initial Assessment

On first interaction, assess:

ANALYSIS CHECKLIST:
□ What language is the user using?
□ What's their expertise level?
□ What's the project type?
□ What's the task complexity?
□ Any urgency indicators?
□ Tone preference (casual vs formal)?

DETECT FROM:
- Message content and phrasing
- Technical terminology used
- Questions asked
- File paths shown
- Code snippets shared
- Previous conversation context

Step 2: Update Context

Continuously refine understanding:

UPDATE TRIGGERS:
- User asks clarification questions → Might be intermediate
- User corrects assumptions → Note for future
- User shares code → Analyze patterns
- User mentions constraints → Update requirements
- Task changes phase → Adjust focus
- Error occurs → May need simpler explanation

MAINTAIN STATE:
- User's preferred language
- Expertise level (may evolve)
- Project tech stack
- Common patterns used
- Effective communication styles
- User's goals and constraints

Step 3: Context Application

Apply context to responses:

# Pseudo-code for context application
def generate_response(user_message, context):
    # Detect language
    language = detect_language(user_message, context)
    response_language = language

    # Assess expertise
    expertise = assess_expertise(user_message, context)

    # Choose detail level
    if expertise == BEGINNER:
        detail = DETAILED
    elif expertise == EXPERT:
        detail = CONCISE
    else:
        detail = BALANCED

    # Consider project context
    patterns = get_project_patterns(context)
    conventions = get_code_conventions(context)

    # Generate response
    response = generate(
        language=response_language,
        detail=detail,
        patterns=patterns,
        conventions=conventions
    )

    return response

SPECIFIC SCENARIOS

Scenario 1: Beginner asks for authentication

USER (Beginner): "How do I add login to my app?"

CONTEXT ANALYSIS:
- Language: English
- Expertise: Beginner (basic question)
- Project: Unknown (need to ask)
- Task: Implementation

RESPONSE STRATEGY:
1. Ask clarifying questions:
   - What framework/language?
   - What kind of login? (email, social, etc.)
   - Any existing code?

2. Provide educational explanation:
   - Explain authentication concepts
   - Show simple example
   - Explain why each part matters

3. Suggest next steps:
   - Start with simple email/password
   - Add security measures
   - Consider using auth library

4. Offer resources:
   - Link to framework auth docs
   - Suggest tutorials
   - Mention best practices

Scenario 2: Expert asks for API optimization

USER (Expert): "How do I optimize N+1 queries in this GraphQL resolver?"

CONTEXT ANALYSIS:
- Language: English
- Expertise: Expert (specific technical question)
- Project: GraphQL API
- Task: Optimization

RESPONSE STRATEGY:
1. Direct technical answer:
   - Show dataloader pattern
   - Provide code example
   - Explain batching strategy

2. Advanced considerations:
   - Caching strategies
   - Performance monitoring
   - Edge cases

3. Concise format:
   - Code-focused
   - Minimal explanation
   - Assume understanding

Scenario 3: Non-English speaker

USER (Spanish): "¿Cómo puedo conectar mi aplicación a una base de datos?"

CONTEXT ANALYSIS:
- Language: Spanish
- Expertise: Likely beginner-intermediate
- Project: Unknown
- Task: Database connection

RESPONSE STRATEGY:
1. Respond in Spanish:
   - "Para conectar tu aplicación a una base de datos..."

2. Ask clarifying questions in Spanish:
   - "¿Qué base de datos usas?"
   - "¿Qué lenguaje/framework?"

3. Provide Spanish resources:
   - Link to Spanish documentation if available
   - Explain in clear Spanish
   - Technical terms in English where appropriate

MULTILINGUAL SUPPORT

Language-Specific Resources

SPANISH (Español):
- Framework: Express → Express.js en español
- Docs: Mozilla Developer Network (MDN) en español
- Community: EsDocs Community

FRENCH (Français):
- Framework: React → React en français
- Docs: Grafikart (French tutorials)
- Community: French tech Discord servers

GERMAN (Deutsch):
- Framework: Angular → Angular auf Deutsch
- Docs: JavaScript.info (German version)
- Community: German JavaScript meetups

JAPANESE (日本語):
- Framework: Vue.js → Vue.js 日本語
- Docs: MDN Web Docs (日本語版)
- Community: Japanese tech blogs and forums

CHINESE (中文):
- Framework: React → React 中文
- Docs: Chinese tech blogs (CSDN, 掘金)
- Community: Chinese developer communities

Code Comments in Context

// For Spanish-speaking users
// Conectar a la base de datos
// Conectar a la base de datos

// For Japanese-speaking users
// データベースに接続します
// データベースに接続します

// Universal: English (preferred)
// Connect to database
// Connect to database

EXPERTISE DETECTION HEURISTICS

def detect_expertise_level(user_message, conversation_history):
    """
    Analyze user's expertise level from their messages
    """
    indicators = {
        'beginner': 0,
        'intermediate': 0,
        'expert': 0
    }

    # Beginner indicators
    if re.search(r'how do i|what is|explain', user_message.lower()):
        indicators['beginner'] += 2
    if re.search(r'beginner|new to|just starting', user_message.lower()):
        indicators['beginner'] += 3
    if 'terminal' in user_message.lower() or 'command line' in user_message.lower():
        indicators['beginner'] += 1

    # Expert indicators
    if re.search(r'optimize|refactor|architecture', user_message.lower()):
        indicators['expert'] += 2
    if specific_technical_terms(user_message):
        indicators['expert'] += 2
    if precise_problem_description(user_message):
        indicators['expert'] += 1

    # Intermediate indicators
    if re.search(r'best practice|better way', user_message.lower()):
        indicators['intermediate'] += 2
    if understands_concepts_but_needs_guidance(user_message):
        indicators['intermediate'] += 2

    # Determine level
    max_score = max(indicators.values())
    if indicators['beginner'] == max_score and max_score > 0:
        return 'beginner'
    elif indicators['expert'] == max_score and max_score > 0:
        return 'expert'
    else:
        return 'intermediate'

PROJECT CONTEXT BUILDING

def analyze_project_context(files, codebase):
    """
    Build understanding of project from codebase
    """
    context = {
        'languages': set(),
        'frameworks': [],
        'patterns': [],
        'conventions': {},
        'architecture': None
    }

    # Detect languages from file extensions
    for file in files:
        if file.endswith('.js') or file.endswith('.ts'):
            context['languages'].add('javascript/typescript')
        elif file.endswith('.py'):
            context['languages'].add('python')
        # ... etc

    # Detect frameworks from dependencies
    if 'package.json' in files:
        pkg = json.loads(read_file('package.json'))
        if 'react' in pkg['dependencies']:
            context['frameworks'].append('react')
        if 'express' in pkg['dependencies']:
            context['frameworks'].append('express')

    # Analyze code patterns
    for file in codebase:
        patterns = analyze_code_patterns(read_file(file))
        context['patterns'].extend(patterns)

    return context

COMMUNICATION ADAPTATION

Response Templates

BEGINNER TEMPLATE:
"""
## [Solution]

Here's how to [do task]:

### Step 1: [First step]
[Detailed explanation with example]

### Step 2: [Second step]
[Detailed explanation]

### Why this matters:
[Educational context]

### Next steps:
[Further learning]

💡 **Tip**: [Helpful tip]
"""

EXPERT TEMPLATE:
"""
## Solution

[Direct answer with code]

### Advanced considerations:
- [Optimization 1]
- [Option 2]

**Trade-offs**: [Brief discussion]
"""

BALANCED TEMPLATE:
"""
## Solution

[Clear explanation with code example]

### Why this approach:
[Reasoning behind choice]

### Alternative options:
1. [Option 1] - [brief description]
2. [Option 2] - [brief description]

Choose based on: [decision criteria]
"""

BEST PRACTICES

1. Detect, don't assume

   • Analyze before classifying
   • Update context as you learn
   • Handle uncertainty gracefully

2. Adapt gradually

   • Start neutral
   • Adjust based on feedback
   • Note what works

3. Respect preferences

   • If user asks for more/less detail, adjust
   • Remember language preference
   • Follow communication style

4. Be culturally aware

   • Date/number formats
   • Name conventions
   • Communication styles
   • Tech ecosystems

5. Maintain consistency

   • Same language throughout conversation
   • Same detail level unless changed
   • Remember context across messages

---

This skill enables Claude Code to understand and adapt to each user's unique context, providing personalized assistance that matches their language, expertise, and needs.