Command Builder for Claude Code FAQs

Question 1

How do I add real-time project context to a command?

Accepted Answer

You can use the exclamation mark prefix followed by backticks (e.g., ! `git status`) to execute bash commands. Their output is automatically injected into the prompt when the command is run.

Question 2

Can I pass variables or arguments to my custom commands?

Accepted Answer

Yes, you can use $ARGUMENTS to capture all input, or positional variables like $1, $2, and $3 for specific structured inputs defined in the command's YAML frontmatter.

Question 3

Where should I save my command files?

Accepted Answer

Project-specific commands should be placed in a 'commands/' directory within your project, while personal commands available across all projects should be saved in '~/commands/'.

Question 4

What is a Claude Code slash command?

Accepted Answer

A slash command is a specialized markdown file that defines a reusable prompt. When you type /command-name in Claude Code, it expands into a detailed set of instructions and context for the AI.

Question 5

How do I restrict the tools Claude can use for a command?

Accepted Answer

Use the 'allowed-tools' field in the YAML frontmatter. You can specify single tools, arrays of tools, or even restrict Bash commands to specific patterns like 'Bash(git add:*)'.

Command Builder for Claude Code

概要

主な機能

ユースケース

Command Builder for Claude Code

概要

主な機能

ユースケース