claude-skills

🎯 65 Skills • 🚀 9 Workflows • 🧠 Context Engineering • 📖 Progressive Disclosure

Quick Start

Get started in minutes with our Quick Start Guide.

TL;DR:

/plugin marketplace add jeffallan/claude-skills

then

/plugin install fullstack-dev-skills@jeffallan

New: Use /common-ground to surface and validate Claude's assumptions about your project before starting work. Add --graph to visualize the reasoning structure as a mermaid diagram.

Note: The Project Workflow Commands require an Atlassian MCP server for Jira and Confluence integration. See the Atlassian MCP Setup Guide for configuration instructions.

Architecture

Progressive Disclosure Pattern

Each skill follows this structure:

skills/react-expert/
├── SKILL.md                    # Lean core (80 lines)
│   ├── Role definition
│   ├── When to use
│   ├── Core workflow
│   └── Routing table          # Points to references
└── references/                 # Loaded on-demand
    ├── server-components.md    # RSC patterns
    ├── react-19-features.md    # use() hook, actions
    ├── state-management.md     # Context, Zustand, Redux
    ├── hooks-patterns.md       # Custom hooks, optimization
    ├── performance.md          # memo, lazy, virtualization
    └── testing-react.md        # Testing Library patterns

How It Works:
1. Skill loads with minimal context (~80 lines)
2. Claude reads the routing table
3. Loads specific references only when context requires
4. 50% faster initial responses, surgical precision when needed

Project Structure

claude-skills/
├── .claude-plugin/
│   ├── plugin.json           # Plugin metadata
│   └── marketplace.json      # Marketplace configuration
├── skills/                   # 65 specialized skills
│   ├── react-expert/
│   │   ├── SKILL.md
│   │   └── references/       # 6 reference files
│   ├── nestjs-expert/
│   │   ├── SKILL.md
│   │   └── references/       # 5 reference files
│   ├── python-pro/
│   │   ├── SKILL.md
│   │   └── references/       # Language-specific patterns
│   └── ... (62 more skills)
├── commands/
│   ├── common-ground/        # Context engineering command
│   │   ├── COMMAND.md
│   │   └── references/
│   └── project/              # 9 project workflow commands
│       ├── discovery/        # Research & validation
│       ├── planning/         # Epic & implementation planning
│       ├── execution/        # Ticket implementation
│       └── retrospectives/   # Reports & completion
├── docs/
│   ├── COMMON_GROUND.md      # Context engineering guide
│   ├── WORKFLOW_COMMANDS.md  # Workflow documentation
│   └── ATLASSIAN_MCP_SETUP.md # MCP server setup guide
├── README.md
├── SKILLS_GUIDE.md          # Quick reference guide
└── CONTRIBUTING.md          # Contribution guidelines

Stats:
- 65 skills
- 351 reference files
- ~50% token reduction
- Covers 30+ frameworks

Skills Overview

65 specialized skills across 12 categories:

• Languages (12): Python Pro, TypeScript Pro, JavaScript Pro, Go Pro, Rust Engineer, SQL Pro, C++ Pro, Swift Expert, Kotlin Specialist, C# Developer, PHP Pro, Java Architect
• Backend Frameworks (7): NestJS Expert, Django Expert, FastAPI Expert, Spring Boot Engineer, Laravel Specialist, Rails Expert, .NET Core Expert
• Frontend & Mobile (6): React Expert, Next.js Developer, Vue Expert, Angular Architect, React Native Expert, Flutter Expert
• Infrastructure (5): Kubernetes Specialist, Terraform Engineer, Postgres Pro, Cloud Architect, Database Optimizer
• API & Architecture (8): GraphQL Architect, API Designer, WebSocket Engineer, Microservices Architect, MCP Developer, Architecture Designer, Feature Forge, Spec Miner
• Testing & Quality (4): Test Master, Playwright Expert, Code Reviewer, Code Documenter
• DevOps & Operations (5): DevOps Engineer, Monitoring Expert, SRE Engineer, Chaos Engineer, CLI Developer
• Security (2): Secure Code Guardian, Security Reviewer
• Data & Machine Learning (6): Pandas Pro, Spark Engineer, ML Pipeline, Prompt Engineer, RAG Architect, Fine-Tuning Expert
• Platform Specialists (4): Salesforce Developer, Shopify Expert, WordPress Pro, Atlassian MCP
• Specialized (3): Legacy Modernizer, Embedded Systems, Game Developer
• Workflow (2): Debugging Wizard, Fullstack Guardian

See SKILLS_GUIDE.md for when to use each skill, workflows, and examples.

Usage Patterns

Context-Aware Activation

Skills activate automatically based on your request:

# Backend Development
"Implement JWT authentication in my NestJS API"
→ Activates: NestJS Expert
→ Loads: references/authentication.md

# Frontend Development
"Build a React component with Server Components"
→ Activates: React Expert
→ Loads: references/server-components.md

# Performance Optimization
"My React app is slow, help me optimize"
→ Activates: React Expert + Debugging Wizard
→ Loads: references/performance.md, references/profiling.md

# Security Review
"Review this authentication code for security issues"
→ Activates: Security Reviewer + Secure Code Guardian
→ Loads: references/auth-patterns.md, references/owasp-top-10.md

Multi-Skill Workflows

Complex tasks combine multiple skills:

Full Feature Development:

Feature Forge → Architecture Designer → Fullstack Guardian → Test Master → Security Reviewer → DevOps Engineer

Bug Investigation:

Debugging Wizard → Framework Expert → Test Master → Code Reviewer

Security Hardening:

Secure Code Guardian → Security Reviewer → Test Master

Context Engineering

/common-ground — Surface Claude's Hidden Assumptions

Claude operates on assumptions about your project—tech stack, coding standards, architecture decisions. This command makes them explicit before they cause misaligned work.

/common-ground              # Surface & validate assumptions interactively
/common-ground --list       # View all tracked assumptions
/common-ground --check      # Quick validation of existing assumptions
/common-ground --graph      # Visualize reasoning structure as mermaid diagram

How it works:

1. Claude analyzes your codebase (config files, code patterns, conversation context)
2. Surfaces assumptions with confidence tiers:
3. ESTABLISHED — High confidence, treat as premises
4. WORKING — Medium confidence, use but flag if contradicted
5. OPEN — Low confidence, ask before assuming
6. You validate, adjust weights, or reject assumptions
7. Claude remembers and respects your decisions across sessions

Example output:

ESTABLISHED: TypeScript strict mode enabled [inferred from tsconfig.json]
WORKING: Prefer functional components over classes [inferred from codebase patterns]
OPEN: Server-side rendering required? [uncertain - needs clarification]

--graph — Visualize Reasoning Structure

The --graph flag generates a mermaid diagram showing the decision tree behind Claude's reasoning—not just what it assumes, but why.

flowchart TD
    ROOT[Task: Build auth system] --> D1{MVP or Production?}
    D1 -->|"0.8 [inferred]"| P1[Production-grade]
    D1 -->|"0.2 [alternative]"| P2[MVP]
    P1 --> D2{Stateless?}
    D2 -->|"0.7 [assumed]"| S1[JWT + refresh]
    D2 -->|"0.3 [uncertain]"| S2[Redis sessions]

Node colors indicate confidence: green (chosen), yellow (decision point), orange (uncertain), gray (alternative not taken).

Project Workflow Commands

Manage your entire development lifecycle with 9 project workflow commands organized into 4 phases:

Commands integrate with Jira (ticket management) and Confluence (documentation publishing).

See docs/WORKFLOW_COMMANDS.md for detailed workflow diagrams, command reference, and integration guides.

Tech Stack Coverage

Languages

• TypeScript / JavaScript
• Python
• Go
• Rust
• C++
• Swift
• Kotlin
• C#
• PHP
• Ruby
• Java
• SQL
• Dart

Backend

• NestJS (TypeScript)
• Django / Django REST Framework (Python)
• FastAPI (Python)
• Express (TypeScript)
• Spring Boot (Java)
• Laravel (PHP)
• Ruby on Rails (Ruby)
• .NET Core (C#)
• GraphQL APIs

Frontend

• React 19 (Server Components, use() hook)
• Next.js (App Router, Server Actions)
• Vue 3 (Composition API, Pinia)
• Angular (Standalone Components, Signals)
• React Native (Expo, bare workflow)
• Flutter (Material Design, Cupertino)

Testing

• Jest / Vitest
• Playwright
• React Testing Library
• Pytest

Databases

• PostgreSQL (TypeORM, Prisma, SQLAlchemy)
• MongoDB (Mongoose, Motor)
• MySQL / MariaDB

Infrastructure & DevOps

• Docker & Docker Compose
• Kubernetes (K8s, Helm)
• Terraform
• AWS / Azure / GCP
• PostgreSQL / MySQL / MongoDB
• CI/CD (GitHub Actions, GitLab CI)

Monitoring

• Prometheus / Grafana
• ELK Stack
• DataDog
• Sentry

Installation Options

Option 1: Marketplace (Recommended)

/plugin marketplace add jeffallan/claude-skills

then

/plugin install fullstack-dev-skills@jeffallan

Option 1a: Installing Skills via npx add-skill

You can install standard Claude skills using the add-skill CLI:

npx add-skill <skill-name>


### Option 2: Local Development

```bash
# Clone the repository
git clone https://github.com/jeffallan/claude-skills.git
cd claude-skills

# Add as local marketplace
/plugin marketplace add /absolute/path/to/claude-skills

# Install from local
/plugin install fullstack-dev-skills@local

Option 3: Direct Installation

# Copy skills directly to Claude Code
cp -r ./skills/* ~/.claude/skills/

Note: Direct installation bypasses plugin management but works for quick testing.

Documentation

• SKILLS_GUIDE.md - Quick reference for when to use each skill
• docs/COMMON_GROUND.md - Context engineering with /common-ground
• docs/WORKFLOW_COMMANDS.md - Project workflow commands guide
• docs/ATLASSIAN_MCP_SETUP.md - Atlassian MCP server setup
• CONTRIBUTING.md - Guidelines for contributing
• skills/*/SKILL.md - Individual skill documentation
• skills/*/references/ - Deep-dive reference materials

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.

Adding a New Skill

1. Create skill directory:
   bash mkdir -p skills/my-skill/references

2. Create lean SKILL.md with routing table:
   ```markdown

name: My Skill
description: Brief description with trigger keywords
triggers:
- keyword1
- keyword2

# My Skill

## Reference Guide
| Topic | Reference | Load When |
|-------|-----------|-----------|
| Topic 1 | references/topic1.md | Context |
```

1. Create reference files (4-6 recommended)

2. Update plugin.json

3. Test locally and submit PR

Adding Reference Files

Keep references focused (200-400 lines each):
- Single topic per file
- Code examples included
- Clear when-to-use guidance
- Cross-references where helpful

Changelog

See CHANGELOG.md for full version history and release notes.

License

MIT License - See LICENSE file for details.

Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Repository: github.com/jeffallan/claude-skills

Author

Built by jeffallan

Principal Consultant at Synergetic Solutions

Fullstack engineering, security compliance, and technical due diligence for teams leveraging AI.

Need help operationalizing AI workflows? Let's talk

:clap: Thanks For Your support

Built for Claude Code | 9 Workflows | 351 Reference Files | 65 Skills