nodnarbnitram__claude-code-extensions/plugins/cce-core

CCE Core Plugin

Essential Claude Code extensions providing core agents, hooks, commands, and universal tools.

Overview

The cce-core plugin is the foundational plugin that provides essential automation, safety hooks, and universal development agents. Everyone should install this plugin as it contains the core functionality needed for safe and productive Claude Code usage.

Features

• 8 Lifecycle Hooks: Safety guards, logging, context loading, and automation
• 13 Core Agents: Code reviewers, orchestrators, meta-agents, and universal specialists
• 9 Slash Commands: Git workflows, project analysis, security scanning, and extension scaffolding
• 2 Essential Skills: Commit message generation and code review automation

Plugin Components

Agents (13)

Core Agents:

• code-archaeologist: Explore unfamiliar/legacy codebases
• code-reviewer: Rigorous security-aware code review
• documentation-specialist: Project documentation creation
• performance-optimizer: Performance analysis and optimization

Orchestrators:

• tech-lead-orchestrator: Strategic project planning
• project-analyst: Codebase analysis and tech stack detection
• team-configurator: AI team setup for projects

Universal Agents:

• api-architect: RESTful/GraphQL API design
• frontend-developer: Framework-agnostic UI development
• backend-developer: Server-side development across stacks
• tailwind-css-expert: Tailwind CSS styling
• fact-checker: Verification and accuracy validation
• meta-agent: Generate new Claude Code agents

Skills (2)

• commit-helper: Smart commit message generation
• code-reviewer: Automated code quality checks

Commands (9)

• /cce:git-commit: Intelligent git commit workflow
• /cce:git-status: Repository state analysis
• /cce:prime: Load project context
• /cce:agent-from-docs: Create agents from documentation URLs
• /cce:agent-intent-from-docs: Create specialized agents with specific intent
• /cce:security-scan: Run security scans on code
• /cce:wrapup-skillup: Generate session learning reports
• /cce:frontend-mode: Load frontend development rules
• /cce:create-skill: Scaffold a new skill from the template structure

Hooks (8 Lifecycle Events)

• PreToolUse: Block dangerous commands, prevent .env commits
• PostToolUse: Log tool executions
• SessionStart: Load project context
• UserPromptSubmit: Mandatory skill activation
• Stop: Session cleanup
• SubagentStop: Subagent completion tracking
• PreCompact: Context compaction logging
• Notification: System notifications

Installation

From Marketplace (Recommended)

# Add the CCE marketplace
/plugin marketplace add github:nodnarbnitram/claude-code-extensions

# Install core plugin
/plugin install cce-core@cce-marketplace

From Local Source

git clone https://github.com/nodnarbnitram/claude-code-extensions.git
/plugin marketplace add /path/to/claude-code-extensions
/plugin install cce-core@cce-marketplace

Usage

Commands (Namespaced)

/cce:git-commit           # Smart commit with conventional format
/cce:prime               # Load project context and structure
/cce:security-scan       # Scan for security vulnerabilities

Agents (Automatic Activation)

> Review this code for security issues
# Automatically uses code-reviewer agent

> Create an agent for working with GraphQL
# Automatically uses meta-agent

> Optimize the slow database queries
# Automatically uses performance-optimizer

Requirements

• Claude Code: Latest version
• Python: 3.11+ (for hooks)
• uv: Python package manager (for hook dependencies)

Safety Features

The core plugin includes critical safety hooks:

✅ Blocks dangerous rm commands ✅ Prevents .env file commits ✅ Logs all tool executions ✅ Validates security-sensitive operations

License

MIT License - see LICENSE for details.

Support

• Issues: GitHub Issues
• Documentation: Repository README