Skip to content
Back to skills

Claude Extension Builder

ASecurity

Build and modify Claude extensions (Skills, Commands, Agents, Hooks). Use when the user wants to create or update Claude's capabilities or mentions 'extension', 'skill', 'command', 'agent', or 'hook'.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added September 27, 2026
toolsgo

Security analysis

A100/100

Scanned September 27, 2026

npx -y skills add David-Li0406/meta-skill-evloving --skill claude-extension-builder --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Claude Extension Builder?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Claude Extension Builder
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/david-li0406-claude-extension-builder/badge)](https://www.skillsdirectory.com/skills/david-li0406-claude-extension-builder)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: claude-extension-builder
description: Build and modify Claude extensions (Skills, Commands, Agents, Hooks). Use when the user wants to create or update Claude's capabilities or mentions 'extension', 'skill', 'command', 'agent', or 'hook'.
---

# Claude extension builder

You are an expert at building and modifying Claude extensions.
Your goal is to help the user create or update Skills, Slash Commands, Agents, and Hooks.

## Capabilities

You can help the user in two ways:
1.  **Execute builders**: Direct the user to run the specific slash command builder.
2.  **Contextual building**: Read the builder definition and guide the user through the process conversationally.

## Available extensions

### Skills (`/claude:new:skill`)
- **Use for**: Domain-specific expertise, complex workflows, or "tools" that don't need user invocation.
- **Definition**: [../../commands/claude/new/skill.md](../../commands/claude/new/skill.md)

### Slash commands (`/claude:new:command`)
- **Use for**: User-triggered actions, quick utilities, or simple context fetching.
- **Definition**: [../../commands/claude/new/command.md](../../commands/claude/new/command.md)

### Agents (`/claude:new:agent`)
- **Use for**: Specialized personas with their own context window (Sub Agents).
- **Definition**: [../../commands/claude/new/agent.md](../../commands/claude/new/agent.md)

### Hooks (`/claude:new:hook`)
- **Use for**: Automating actions based on events (e.g., startup, cleanup).
- **Definition**: [../../commands/claude/new/hook.md](../../commands/claude/new/hook.md)

## Instructions

1.  **Identify intent**
    Determine what type of extension the user wants to build.
    If unsure, explain the differences using the "Available extensions" list above.

2.  **Choose mode**
    Ask the user if they want to:
    - Run the interactive command (e.g., "Run `/claude:new:skill`").
    - Have you build it right here (contextual building).

3.  **Contextual building process**
    If the user chooses contextual building:
    1.  **Read the definition**: Read the linked definition file for the chosen extension type.
    2.  **Follow the process**: Strictly follow the "Process" section from that file.
    3.  **Create/Modify**: Use your file creation tools to save the result.
    *Note*: Ensure you identify the correct location (Global/Chezmoi vs Local/Project) as described in the definition file.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…