Agent Skills

cloudflare-manager

Professional slash commands for Claude Code that provide structured workflows for software development tasks including code review, feature creation, security auditing, and architectural analysis.

Installation
CLI
npx skills add https://github.com/qdhenry/Claude-Command-Suite --skill cloudflare-manager

Installez cette compétence avec la CLI et commencez à utiliser le flux de travail SKILL.md dans votre espace de travail.

Dernière mise à jour le 5/1/2026

Claude Command Suite

Total Commands
Skills
AI Agents
GitHub Release
License

A comprehensive development toolkit designed following Anthropic's Claude Code Best Practices for AI-assisted software development.

Claude Command Suite is a development toolkit providing 216+ slash commands, 12 Claude Code Skills, 54 AI agents, and automated workflows for software engineering tasks. The suite covers code review, testing, deployment, media processing, WebMCP integration, business scenario modeling, and GitHub-Linear synchronization through structured, repeatable workflows.

  • Slash Commands: Namespace-organized commands (/dev:code-review, /test:generate-test-cases, /deploy:prepare-release) for development workflows
  • AI Agents: Specialized agents for security auditing, test coverage analysis, and cross-platform synchronization
  • Claude Code Skills: 12 model-invoked capabilities for recurring workflows including Linear task management, WebMCP integration, media processing, and Cloudflare deployment
  • Standalone Commands: Purpose-built commands for project cleanup, spec writing, file watching, WebMCP, and media processing
  • Automated Workflows: Pre-configured sequences for feature implementation and production deployment
  • Scenario Simulators: Decision-making tools for business modeling, system behavior analysis, and architectural evaluation
  • Task Orchestration: Task management system with Git integration and context preservation

AI Agents

AI Agents
Workflows

Specialized AI agents handle complex development tasks through focused tool access and isolated contexts:

  • Code Quality Suite: Automated code review, security scanning, performance analysis
  • Test Engineer: Test generation with coverage analysis
  • Integration Manager: GitHub-Linear bidirectional synchronization
  • Strategic Analyst: Business scenario modeling and decision analysis
  • Additional Agents: Project initialization, release management, architecture review

AI Agents Documentation | Workflow Examples

Claude Code Skills

Skills Available
Documentation

Claude Code Skills extend the suite with model-invoked capabilities that complement slash commands. Skills are automatically activated by Claude based on context, suitable for recurring workflows and domain-specific expertise.

Skills vs Commands

Feature Skills (Model-Invoked) Commands (User-Invoked)
Activation Contextually triggered Explicit invocation via /command
Use Case Recurring workflows, domain expertise Specific task execution
Scope Personal or project-wide Project-specific
Distribution Git repositories, plugins Command file copying

Available Skills

linear-todo-sync - Linear task management integration

  • GraphQL API integration for task retrieval
  • Priority-based organization with metadata
  • Markdown output with actionable links
  • Context triggers: "What do I need to work on?", "Show my work"

cloudflare-manager - Cloudflare infrastructure management

  • Workers deployment with URL extraction
  • KV Storage and R2 bucket management
  • Cloudflare Pages deployment
  • DNS and routing configuration
  • Requirements: Bun runtime, CLOUDFLARE_API_KEY

webmcp - WebMCP browser-native AI tool integration

  • Imperative JavaScript API and declarative HTML form annotation
  • Tool setup, debugging, auditing, and testing workflows
  • Chrome 146+ with WebMCP flag enabled
  • Model Context Tool Inspector integration

bigcommerce-api - BigCommerce API expert

  • REST V2/V3, GraphQL Storefront, webhooks, and OAuth
  • Headless storefront development with Next.js/Catalyst
  • Catalog, Orders, Customers, and Payments API workflows
  • Multi-storefront channel awareness and rate limit management

audit-env-variables - Environment variable security audit

  • Scans .env* files and cross-references usage across JS/TS codebases
  • Detects unused variables, dynamic access patterns, and known services
  • Optional cleanup mode with backup branches and regression prevention

remove-dead-code - Multi-agent dead code removal

  • Parallel scout agents analyze unused exports, orphaned files, dead imports
  • Automatic backup branch creation before any removal
  • Post-removal validation with TypeScript checking, build, and test suite

elevenlabs-transcribe - Audio/video transcription

  • ElevenLabs Scribe v2 API with speaker diarization (up to 32 speakers)
  • Audio event tagging, language detection, and key-term biasing
  • Requirements: ELEVENLABS_API_KEY, uv runtime

extract-video-frames - Video frame extraction

  • PNG frame extraction at configurable intervals via ffmpeg
  • Aligned audio segments and full audio track export
  • JSON manifest generation for agent handoff pipelines

file-watcher - Chokidar-based file change watcher

  • Monitors files/directories and triggers claude -p on changes
  • Glob filtering, debounce control, and {{file}} placeholder support
  • Use cases: auto code review, type checking, test generation

gsap-animation - GSAP animation reference

  • Complete API reference covering tweens, timelines, easing, stagger, and plugins
  • ScrollTrigger, Flip, SplitText, Draggable deep references
  • React integration with useGSAP() hook and SSR/Next.js patterns

setup-agent-tail - Agent-tail log aggregation setup

  • Auto-detects framework (Vite, Next.js, monorepo, Node CLI)
  • Browser console capture and configurable log filtering
  • Package manager detection and script wrapping

setup-portless - Portless named .localhost URLs

  • Replaces port numbers with stable name.localhost URLs
  • Monorepo subdomain naming patterns
  • Requirements: Node.js 20+, macOS or Linux

Skills Builder Framework

Structured workflow for creating custom skills:

  • /skills:build-skill - Guided skill creation command
  • Four-phase process: Requirements elicitation, code generation, validation, documentation
  • Quality assurance: Automated testing and validation scripts
  • Templates: Five skill templates for different use cases

Usage

Triggering skills:

"What do I need to work on today?"  # Activates linear-todo-sync
"Deploy a cloudflare worker"        # Activates cloudflare-manager
"Set up WebMCP in this project"     # Activates webmcp
"Transcribe this audio file"        # Activates elevenlabs-transcribe
"Find and remove dead code"         # Activates remove-dead-code

Creating skills:

/skills:build-skill

Documentation:

📚 Table of Contents

🤖 AI Agents 🎨 Skills 🚀 Getting Started ⚡ Commands 💡 Usage
AI Agents Skills Quick Start Available Commands How It Works
Agent Docs Skills Quickstart Installation Guide Command Namespaces Common Workflows
Workflows Skills Docs Contributing

Quick Start

Simply add command files directly to your project's .claude/commands/ folder:

# Create the commands directory in your project
mkdir -p .claude/commands

# Download specific commands you need
curl -o .claude/commands/dev/code-review.md https://raw.githubusercontent.com/qdhenry/Claude-Command-Suite/main/.claude/commands/dev/code-review.md

# Or copy your own custom command files
cp my-custom-command.md .claude/commands/

Option 2: Install All Commands

Use our installation script to set up all commands at once:

git clone https://github.com/qdhenry/Claude-Command-Suite.git
cd Claude-Command-Suite
chmod +x install.sh
./install.sh

Start Using Commands

claude code
/dev:code-review           # Review your entire codebase
/project:create-feature    # Build a new feature
/security:security-audit   # Check for security issues
/dev:cleanup-vibes         # Clean up a vibecoded project
/webmcp:setup              # Set up WebMCP integration

Available Commands

216 commands organized by namespace for software development workflows.

54 AI agents available for complex task automation. AI Agents Documentation

Command Namespace Index

Core Development Quality & Testing Deployment Collaboration Advanced
🚀 Project 🧪 Test 📦 Deploy 👥 Team 🎯 Simulation
💻 Dev 🔒 Security 📚 Docs 🔄 Sync 📋 Orchestration
🔧 Setup ⚡ Performance 🌐 WebMCP 💬 Session 🧠 WFGY
🦀 Rust 🎬 Media

Command Namespaces

/project:* - Project Management

Project initialization, configuration, and management. Commands for project creation, dependency management, milestone tracking, and health monitoring.

/dev:* - Development Tools

Development utilities including code review, debugging, refactoring, and analysis modes.

/test:* - Testing Suite

Testing tools for unit tests, integration tests, E2E tests, coverage analysis, mutation testing, and visual regression testing.

/security:* - Security & Compliance

Security auditing, dependency scanning, authentication implementation, and security hardening for codebase security.

/performance:* - Performance Optimization

Tools for optimizing build times, bundle sizes, database queries, caching strategies, and application performance.

/sync:* - Integration & Synchronization

Bidirectional synchronization between GitHub Issues and Linear, PR tracking, conflict resolution, and cross-platform task management.

/deploy:* - Deployment & Release

Release preparation, automated deployments, rollback capabilities, containerization, and Kubernetes deployment management.

/docs:* - Documentation Generation

Documentation automation for APIs, architecture diagrams, onboarding guides, and troubleshooting documentation.

/setup:* - Configuration & Setup

Setup commands for development environments, linting, formatting, monitoring, database schemas, and API design.

/team:* - Team Collaboration

Team workflow tools including standup reports, sprint planning, retrospectives, workload balancing, and knowledge management.

/simulation:* - Scenario Simulators

Inspired by "AI agents at their most under-leveraged point" by AI News & Strategy Daily | Nate B. Jones

Simulation and modeling tools for decision analysis through scenario exploration, digital twins, and timeline compression.

📖 Comprehensive Examples Guide - Real-world usage patterns, argument templates, and command synergy workflows.

/rust:* - Rust & Tauri Development

Rust codebase auditing, refactoring, and Tauri desktop application development tools.

Architecture Auditing:

Tauri Desktop App (/rust:tauri:*):

/webmcp:* - WebMCP Integration

Browser-native AI tool integration using Chrome's WebMCP standard. Setup, tool creation, debugging, and auditing workflows.

/media:* - Media Processing

Audio and video processing tools for frame extraction, transcription, and agent pipeline handoff.

/session:* - Session Management

Context handoff and session continuity tools for preserving work across Claude sessions.

  • /session:handoff - Create comprehensive handoff documents capturing work context, decisions, and next steps
  • /session:handoff-continue - Create handoff document and automatically spawn a new Claude session in a Zellij pane

Development tools /dev:prime, /dev:all-tools contributed by IndyDevDan (YouTube: https://www.youtube.com/@indydevdan) / DislerH (GitHub: https://github.com/disler)

/orchestration:* - Task Orchestration

Task management and execution system for organizing complex projects into trackable workflows. Task decomposition, progress tracking, Git synchronization, and context preservation.

📚 Task Orchestration Guide - Complete guide with examples, workflows, and best practices for managing complex development projects.

/wfgy:* - Semantic Reasoning & Memory

Semantic reasoning system based on the WFGY project providing mathematical validation, persistent memory, and hallucination prevention.

Core Formula Commands (/wfgy:*):

Semantic Memory (/semantic:*):

Knowledge Boundaries (/boundary:*):

Reasoning Operations (/reasoning:*):

Memory Management (/memory:*):

📚 WFGY Documentation - Mathematical foundations, performance benchmarks, and comprehensive usage guide for semantic reasoning.

How It Works

Claude Code automatically recognizes markdown files in .claude/commands/ as slash commands. Files are loaded on startup and available immediately.

Command Structure

Each command file needs:

  • Title: # Command Name
  • Description: Brief overview of what the command does
  • Instructions: ## Instructions section with detailed steps

Example minimal command (my-command.md):

# My Custom Command

Performs a specific task in my project.

## Instructions

1. **First Step**
   - Do this thing
   - Check that thing

2. **Second Step**
   - Execute this action
   - Verify the result

Use it immediately with: /namespace:my-command

Using Commands

Commands use the format /namespace:command-name. Each command executes a structured workflow defined in its markdown file.

Examples:

  • /dev:code-review - Analyze codebase for quality, security, and performance
  • /project:create-feature dashboard - Plan, implement, and test a new feature
  • /dev:fix-issue 123 - Resolve GitHub issue with systematic approach
  • /simulation:business-scenario-explorer Evaluate market expansion - Model business scenarios

Common Workflows

New Feature Development:

/dev:code-review                    # Assess current state
/project:create-feature user-dashboard  # Implement feature
/security:security-audit                 # Verify security

Bug Fixing:

/dev:fix-issue 456                  # Fix the specific issue
/dev:code-review                    # Verify fix quality

Code Maintenance:

/security:dependency-audit               # Check for outdated dependencies
/performance:performance-audit              # Identify bottlenecks
/dev:refactor-code legacy-module    # Improve problematic areas

Project Cleanup & Spec Writing:

/dev:cleanup-vibes                          # Transform vibecoded project into structured codebase
/spec-workflow:quick-spec auth system       # Rapid spec with opinionated recommendations
/dev:remove-dead-code scan                  # Find dead code across the project

Media Processing Pipeline:

/media:extract-video-frames demo.mp4 2      # Extract frames every 2 seconds
/media:elevenlabs-transcribe interview.mp3   # Transcribe with speaker diarization

WebMCP Integration:

/webmcp:setup                               # Set up WebMCP from scratch
/webmcp:add-tool "search products"          # Add a new tool
/webmcp:audit                               # Audit for best practices

Session Handoff:

/session:handoff                            # Create handoff document for context transfer
/session:handoff-continue                   # Handoff + spawn new Claude session in Zellij

Strategic Decision Making:

/simulation:constraint-modeler              # Map decision constraints
/simulation:business-scenario-explorer      # Explore multiple timelines
/simulation:decision-tree-explorer          # Optimize decision choices

Complex Project Management:

/orchestration:start                        # Break down project into tasks
/orchestration:status                       # Monitor progress
/orchestration:resume                       # Continue after breaks
/orchestration:commit                       # Create linked Git commits

Additional Documentation

Contributing

This repository accepts contributions for slash commands, AI agents, and skills.

Contribution Areas

  • Commands - New workflow commands or enhancements to existing ones
  • AI Agents - Specialized agents for development tasks
  • Skills - Model-invoked capabilities for recurring workflows
  • Documentation - Improvements to guides and examples
  • Bug Fixes - Issues with existing functionality

Contribution Process

  1. Fork the repository
  2. Add or modify content in appropriate directories
  3. Test changes according to guidelines
  4. Submit pull request with clear description

See Development Guide for contribution guidelines, testing requirements, and quality standards.

Automated Changelog

Changelog generation is automated through GitHub Actions and manual scripts.

Automatic Updates

  • GitHub Actions integration detects command additions
  • Tracks new files and modifications
  • Generates formatted changelog entries
  • Adds summary comments to pull requests

Manual Updates

Run the update script for manual changelog generation:

./scripts/update-changelog.sh

The script performs the following operations:

  • Scans command files in .claude/commands/
  • Extracts command names and descriptions
  • Updates CHANGELOG.md with categorized listings
  • Creates backup of existing changelog

Features

  • Command categorization by type
  • Total count tracking
  • Description extraction from source files
  • Backup creation before modifications

Release Process

Automated releases use Conventional Commits. Merges to main trigger the release workflow:

  1. Commit analysis determines version bump type
  2. Version updates based on commit types
  3. Release notes generated from commit messages
  4. GitHub releases created with downloadable bundles
  5. Changelog updated with categorized changes

Version Bumping

  • fix: commits → Patch release (1.0.0 → 1.0.1)
  • feat: commits → Minor release (1.0.0 → 1.1.0)
  • Breaking changes → Major release (1.0.0 → 2.0.0)

Release Triggers

  • Pull request merges with conventional commit titles
  • Direct pushes to main with conventional commits

For commit format details, see CONTRIBUTING.md.

Requirements

  • Claude Code version 1.0 or later
  • Commands work with any programming language or framework

Attribution

External AI Agents

This project includes 44 specialized AI agents from wshobson/agents, expanding our collection to 55 total agents. These external agents provide expertise in:

  • Development: Backend architecture, frontend development, mobile development, GraphQL
  • Languages: Python, Go, Rust, C/C++, JavaScript, SQL specialists
  • Infrastructure: DevOps, cloud architecture, database administration, Terraform
  • Quality: Code review, security auditing, performance engineering, test automation

Full attribution and details can be found in .claude/agents/external/wshobson/ATTRIBUTION.md.

Inspired by Anthropic's Claude Code Best Practices