@agentlee5/agent-skills 1.0.0

package/skills/agent-patterns/agent-design-patterns/SKILL.md

@@ -0,0 +1,70 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: AI.AGENT.PATTERNS
+TAG: AI.AGENT.PATTERNS.AGENT_DESIGN_PATTERNS
+
+COLOR_ONION_HEX:
+NEON=#00FFFF
+FLUO=#00CED1
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = agent design patterns skill for Leeway-compliant AI systems
+WHY = Provides capabilities for agent-patterns within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/agent-patterns/agent-design-patterns/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Agent Patterns and Architectures
+
+**Expert in**: Building scalable, maintainable agent systems using proven design patterns and best practices.
+
+## Capabilities
+
+- Implement ReAct (Reasoning + Acting) pattern
+- Design tool-using agents with autonomous decision-making
+- Create planning agents with goal decomposition
+- Build reflection and self-improvement mechanisms
+- Implement hierarchical reasoning (System 1 / System 2)
+- Design agentic loops with memory and state
+- Create specialized agent roles and personas
+- Implement agent memory systems (short-term, long-term)
+
+## Use this skill when:
+
+- Building autonomous agents that use tools
+- Implementing complex reasoning workflows
+- Creating agents that improve over time
+- Designing agent systems with specialized roles
+- Building agents that plan and execute multi-step tasks
+- Implementing agent-based research systems
+- Creating agents with persistent memory
+
+## Key techniques
+
+- ReAct loops (Thought → Action → Observation)
+- Tool calling and function composition
+- Goal tree decomposition
+- Monte Carlo tree search for planning
+- Iterative refinement and reflection
+- Memory-augmented agents
+- Agent personas and role-based behavior
+- Skill stacking and tool composition
+
+## Tags
+
+`agents` `patterns` `reasoning` `tool-use` `planning` `autonomous`

package/skills/ai-ml/llm-prompting/SKILL.md

@@ -0,0 +1,71 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: AI.SKILL.AI_ML
+TAG: AI.SKILL.AI_ML.LLM_PROMPTING
+
+COLOR_ONION_HEX:
+NEON=#39FF14
+FLUO=#0DFF94
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = llm prompting skill for Leeway-compliant AI systems
+WHY = Provides capabilities for ai-ml within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/ai-ml/llm-prompting/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# LLM Prompting and Engineering
+
+**Expert in**: Writing effective prompts, prompt engineering, and working with large language models to achieve consistent, high-quality outputs.
+
+## Capabilities
+
+- Write clear, effective prompts with proper instruction structuring
+- Structure complex prompts with examples and demonstrations
+- Use few-shot and zero-shot techniques appropriately
+- Implement chain-of-thought and step-by-step reasoning
+- Create system prompts and instructions with proper role-setting
+- Handle edge cases, errors, and failure modes
+- Optimize for specific models and model behavior
+- Evaluate prompt effectiveness with metrics
+- Design prompt templates and parameterization
+- Implement in-context learning strategies
+- Handle prompt injection and security concerns
+- Create multi-turn conversation flows
+
+## Use this skill when:
+
+- Improving LLM output quality
+- Creating prompt templates
+- Handling complex tasks
+- Getting consistent results
+- Debugging model behavior
+- Optimizing token usage
+
+## Key techniques
+
+- Prompt structure and clarity
+- Few-shot and zero-shot prompting
+- Chain-of-thought reasoning
+- Role-based prompts
+- Output formatting and constraints
+- Temperature and parameter tuning
+
+## Tags
+
+`llm` `prompting` `ai` `engineering` `gpt`

package/skills/ai-ml/ml-model-development/SKILL.md

@@ -0,0 +1,67 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: AI.SKILL.AI_ML
+TAG: AI.SKILL.AI_ML.ML_MODEL_DEVELOPMENT
+
+COLOR_ONION_HEX:
+NEON=#39FF14
+FLUO=#0DFF94
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = ml model development skill for Leeway-compliant AI systems
+WHY = Provides capabilities for ai-ml within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/ai-ml/ml-model-development/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Machine Learning Model Development
+
+**Expert in**: Building, training, and evaluating machine learning models with scikit-learn, TensorFlow, and PyTorch.
+
+## Capabilities
+
+- Prepare and preprocess training data
+- Build classification and regression models
+- Handle imbalanced datasets
+- Feature engineering and selection
+- Model evaluation and validation
+- Hyperparameter tuning
+- Handle overfitting and underfitting
+- Deploy ML models
+
+## Use this skill when:
+
+- Building classification/regression models
+- Preparing ML datasets
+- Feature engineering
+- Model evaluation and selection
+- Hyperparameter optimization
+- Deploying models to production
+
+## Key techniques
+
+- Data preprocessing and normalization
+- Train/validation/test splits and cross-validation
+- Feature scaling and encoding
+- Scikit-learn pipelines
+- Model evaluation metrics
+- GridSearchCV and cross-validation
+
+## Tags
+
+`machine-learning` `scikit-learn` `tensorflow` `pytorch` `data-science`

package/skills/ai-ml/multimodal-systems/SKILL.md

@@ -0,0 +1,71 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: AI.SKILL.AI_ML
+TAG: AI.SKILL.AI_ML.MULTIMODAL_SYSTEMS
+
+COLOR_ONION_HEX:
+NEON=#39FF14
+FLUO=#0DFF94
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = multimodal systems skill for Leeway-compliant AI systems
+WHY = Provides capabilities for ai-ml within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/ai-ml/multimodal-systems/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Multimodal Systems
+
+**Expert in**: Building systems that work with multiple modalities (text, images, audio, video) for comprehensive AI understanding.
+
+## Capabilities
+
+- Integrate vision models for image understanding
+- Build text-to-image generation systems
+- Implement audio processing and transcription
+- Create video understanding systems
+- Design cross-modal retrieval
+- Build multimodal RAG systems
+- Implement multimodal fusion strategies
+- Handle modality-specific challenges
+
+## Use this skill when:
+
+- Building systems understanding images
+- Implementing audio/speech processing
+- Creating generative image systems
+- Processing video content
+- Cross-modal search and retrieval
+- Multimodal document understanding
+- Building comprehensive AI systems
+- Accessibility through multiple modalities
+
+## Key techniques
+
+- Vision transformers (ViT)
+- CLIP and cross-modal embedding
+- Diffusion models for generation
+- Audio embeddings and processing
+- Video understanding and key frames
+- DALL-E, Stable Diffusion, Midjourney APIs
+- Multimodal transformers
+- Modality alignment and fusion
+
+## Tags
+
+`multimodal` `vision` `audio` `video` `image-generation` `cross-modal`

package/skills/ai-ml/retrieval-generation-fine-tuning/SKILL.md

@@ -0,0 +1,71 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: AI.SKILL.AI_ML
+TAG: AI.SKILL.AI_ML.RETRIEVAL_GENERATION_FINE_TUNING
+
+COLOR_ONION_HEX:
+NEON=#39FF14
+FLUO=#0DFF94
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = retrieval generation fine tuning skill for Leeway-compliant AI systems
+WHY = Provides capabilities for ai-ml within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/ai-ml/retrieval-generation-fine-tuning/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Model Fine-Tuning and Adaptation
+
+**Expert in**: Adapting and fine-tuning language models for specific domains and use cases.
+
+## Capabilities
+
+- Prepare training data for domain adaptation
+- Implement LoRA (Low-Rank Adaptation) fine-tuning
+- Create instruction tuning datasets
+- Implement few-shot in-context learning
+- Design domain-specific vocabularies
+- Handle catastrophic forgetting in fine-tuning
+- Evaluate fine-tuned model performance
+- Implement model merging strategies
+
+## Use this skill when:
+
+- Models need domain-specific knowledge
+- Building specialized AI systems
+- Wanting to preserve base model knowledge
+- Reducing hallucinations in specific domains
+- Adapting commercial models
+- Creating proprietary model variants
+- Improving task-specific performance
+- Working with limited labeled data
+
+## Key techniques
+
+- LoRA and QLoRA adaptation
+- Instruction tuning and RLHF
+- Few-shot prompting for adaptation
+- Embedding fine-tuning
+- Continued pretraining
+- Model merging (TIES, linear interpolation)
+- Evaluation metrics for fine-tuning
+- Cost-effective adaptation
+
+## Tags
+
+`fine-tuning` `llm` `adaptation` `lora` `domain-specific` `training`

package/skills/architecture/system-design/SKILL.md

@@ -0,0 +1,67 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: AI.SKILL.ARCHITECTURE
+TAG: AI.SKILL.ARCHITECTURE.SYSTEM_DESIGN
+
+COLOR_ONION_HEX:
+NEON=#39FF14
+FLUO=#0DFF94
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = system design skill for Leeway-compliant AI systems
+WHY = Provides capabilities for architecture within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/architecture/system-design/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# System Design and Architecture
+
+**Expert in**: Designing scalable, maintainable systems and architectural patterns.
+
+## Capabilities
+
+- Design distributed systems
+- Plan microservices architectures
+- Database design and schema
+- API design and REST principles
+- Caching strategies
+- Message queues and event-driven systems
+- Scalability and performance planning
+- System reliability and resilience
+
+## Use this skill when:
+
+- Planning new systems
+- Architectural reviews
+- Performance optimization
+- Scalability planning
+- Database design
+- Component interaction design
+
+## Key techniques
+
+- Design patterns (Factory, Singleton, Observer, etc.)
+- SOLID principles
+- Database normalization and denormalization
+- CAP theorem and consistency models
+- Load balancing and sharding
+- Disaster recovery planning
+
+## Tags
+
+`architecture` `system-design` `distributed-systems` `scalability`

package/skills/code-analysis/refactoring/SKILL.md

@@ -0,0 +1,64 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: DEV.ANALYSIS
+TAG: DEV.ANALYSIS.REFACTORING
+
+COLOR_ONION_HEX:
+NEON=#FF1493
+FLUO=#FF69B4
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = refactoring skill for Leeway-compliant AI systems
+WHY = Provides capabilities for code-analysis within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/code-analysis/refactoring/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Refactoring Code
+
+**Expert in**: Restructuring code to improve readability, maintainability, and performance without changing behavior.
+
+## Capabilities
+
+- Extract methods and classes
+- Simplify complex functions
+- Rename variables and functions for clarity
+- Remove duplication (DRY principle)
+- Consolidate conditional logic
+- Improve module organization
+
+## Use this skill when:
+
+- Code is too complex or hard to understand
+- There's significant duplication
+- Need to improve code organization
+- Preparing for feature additions
+- Simplifying legacy code
+
+## Key techniques
+
+- Extract method/variable patterns
+- Consolidate duplicate code
+- Simplify conditional branches
+- Apply design patterns
+- Module organization
+- Naming improvements
+
+## Tags
+
+`refactoring` `code-improvement` `maintainability` `architecture`

package/skills/code-analysis/security-vulnerability-scanning/SKILL.md

@@ -0,0 +1,71 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: DEV.ANALYSIS
+TAG: DEV.ANALYSIS.SECURITY_VULNERABILITY_SCANNING
+
+COLOR_ONION_HEX:
+NEON=#FF1493
+FLUO=#FF69B4
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = security vulnerability scanning skill for Leeway-compliant AI systems
+WHY = Provides capabilities for code-analysis within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/code-analysis/security-vulnerability-scanning/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Security Vulnerability Scanning
+
+**Expert in**: Identifying security vulnerabilities through automated scanning, analysis, and remediation.
+
+## Capabilities
+
+- Implement SAST (Static Application Security Testing)
+- Deploy DAST (Dynamic Application Security Testing)
+- Manage dependency scanning and CVE tracking
+- Implement container image scanning
+- Create secret detection systems
+- Analyze infrastructure misconfigurations
+- Build security scorecard systems
+- Implement automated remediation workflows
+
+## Use this skill when:
+
+- Finding security vulnerabilities automatically
+- Scanning dependencies for CVEs
+- Compliance and audit requirements
+- Pre-deployment security checking
+- Infrastructure security verification
+- Container security
+- Secret management
+- Continuous security monitoring
+
+## Key techniques
+
+- Semgrep for static analysis
+- OWASP ZAP for dynamic testing
+- Snyk, Dependabot for dependency scanning
+- Trivy for container scanning
+- SonarQube for code quality and security
+- GitGuardian for secret detection
+- Terraform security scanning
+- Supply chain security (SBOM, provenance)
+
+## Tags
+
+`security` `vulnerability` `scanning` `sast` `dast` `compliance`

package/skills/code-analysis/static-analysis/SKILL.md

@@ -0,0 +1,64 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: DEV.ANALYSIS
+TAG: DEV.ANALYSIS.STATIC_ANALYSIS
+
+COLOR_ONION_HEX:
+NEON=#FF1493
+FLUO=#FF69B4
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = static analysis skill for Leeway-compliant AI systems
+WHY = Provides capabilities for code-analysis within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/code-analysis/static-analysis/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Static Code Analysis
+
+**Expert in**: Analyzing code for patterns, anti-patterns, complexity, and potential issues without execution.
+
+## Capabilities
+
+- Identify code smells and anti-patterns
+- Calculate cyclomatic complexity
+- Detect dead code and unused variables
+- Find security vulnerabilities
+- Analyze architectural violations
+- Identify performance anti-patterns
+
+## Use this skill when:
+
+- You need to review code for quality issues
+- Assessing codebase health
+- Finding security vulnerabilities
+- Identifying refactoring opportunities
+- Analyzing architectural decisions
+
+## Key techniques
+
+- AST-based code analysis
+- Metric calculations (complexity, coupling, cohesion)
+- Pattern matching against known issues
+- Dependency graph analysis
+- Security pattern detection
+- Performance analysis
+
+## Tags
+
+`code-analysis` `quality` `security` `performance` `refactoring`

package/skills/code-generation/full-stack-application/SKILL.md

@@ -0,0 +1,70 @@
+/*
+LEEWAY HEADER — DO NOT REMOVE
+
+REGION: DEV.CODEGEN
+TAG: DEV.CODEGEN.FULL_STACK_APPLICATION
+
+COLOR_ONION_HEX:
+NEON=#39FF14
+FLUO=#0DFF94
+PASTEL=#E8F5E9
+
+ICON_ASCII:
+family=lucide
+glyph=zap
+
+5WH:
+WHAT = full stack application skill for Leeway-compliant AI systems
+WHY = Provides capabilities for code-generation within the AIskills ecosystem
+WHO = Leeway Industries (By Leonard Jerome Lee)
+WHERE = skills/code-generation/full-stack-application/SKILL.md
+WHEN = 2026
+HOW = Leeway-governed skill.md definition with structured capabilities and tags
+
+AGENTS:
+ASSESS
+AUDIT
+
+LICENSE:
+MIT
+*/
+
+# Full-Stack Application Delivery
+
+**Expert in**: Building complete, production-ready applications from requirements to deployment, integrating frontend, backend, database, and infrastructure.
+
+## Capabilities
+
+- Design complete application architectures
+- Scaffold full-stack applications with best practices
+- Integrate frontend, API, and data layers
+- Implement authentication and authorization
+- Create end-to-end data flows
+- Package applications for deployment
+- Create comprehensive test coverage
+- Document applications for maintainability
+
+## Use this skill when:
+
+- Building new applications from scratch
+- Need complete end-to-end solutions
+- Creating prototypes rapidly
+- Scaling applications
+- Integrating multiple systems
+- Creating reference implementations
+- Training/onboarding on full-stack patterns
+
+## Key techniques
+
+- Monolithic vs. microservices trade-offs
+- API contract design and versioning
+- Authentication flows (JWT, OAuth2, OIDC)
+- Database migration strategies
+- Frontend-backend communication patterns
+- Testing pyramid implementation
+- Containerization and deployment
+- Infrastructure-application integration
+
+## Tags
+
+`full-stack` `application` `scaffolding` `architecture` `integration` `complete`