slash-command-builder

Design, validate, and troubleshoot Claude Code slash commands with frontmatter and tool permissions.

1|1|Updated Oct 30, 2025
One-click install
npx skills add https://github.com/NangoHQ/ai-ide-prompts --skill slash-command-builder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: slash-command-builder
Source: https://github.com/NangoHQ/ai-ide-prompts/tree/main/.claude/skills/slash-command-builder-skill
Command: npx skills add https://github.com/NangoHQ/ai-ide-prompts --skill slash-command-builder

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Repetitive prompts and common development tasks can slow down workflow. This Skill provides expert guidance for creating effective Claude Code slash commands, enabling users to automate frequent actions, streamline workflows, and boost productivity with simple, explicit invocations.

Core Features & Use Cases

  • Command Structure & Arguments: Learn to design concise command names, define arguments (positional or all), and structure prompts for single-purpose tasks.
  • Frontmatter Configuration: Master description, argument-hint, allowed-tools, model selection, and disable-model-invocation for precise command behavior.
  • Advanced Features: Utilize inline Bash execution (!), file references (@), and specific tool permissions to create powerful yet safe commands.
  • Use Case: A developer frequently performs a specific code review checklist for pull requests. This Skill helps them create a /review-pr command that takes a PR number as an argument, uses Bash(git *) to fetch changes, and then applies a predefined review prompt, saving time on every review.

Quick Start

Use the slash-command-builder skill to help me create a new Claude slash command. I want a command called '/git-status' that shows the current git status and recent commits.

Frequently Asked Questions about slash-command-builder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a custom Claude slash command to automate repetitive tasks?

Design a slash command by defining a concise name, structuring arguments, and writing a focused prompt. Use frontmatter options like `description`, `argument-hint`, and `allowed-tools` to configure behavior, then place the command file in your project or personal command directory for Claude Code to recognize and execute it.

What frontmatter options do I need to configure a Claude slash command?

Essential frontmatter includes `description` (what the command does), `argument-hint` (expected inputs), `allowed-tools` (which tools it can invoke), `model` (which Claude model to use), and `disable-model-invocation` (to run Bash only). These control command visibility, permissions, and execution behavior.

Can I use Bash commands and file references inside a Claude slash command?

Yes. Prefix Bash with `!` (e.g., `!(git status)`) to execute shell commands inline, and use `@` to reference files. This lets you fetch dynamic data like git changes or file contents and pass it directly into your command prompt for processing.

Where do I store Claude slash commands so they're available to use?

Store commands in either your project command directory or personal command directory. Claude Code scans both locations at startup and makes commands matching your frontmatter syntax available as slash invocations.

How do I restrict which tools a slash command can access?

Use the `allowed-tools` frontmatter field to specify tool permissions—for example, `Bash(git *)` limits Bash execution to git commands only. This enforces security boundaries and prevents accidental or unauthorized tool use within the command.

What's the difference between single-purpose and multi-purpose slash commands?

Single-purpose commands execute one focused task with clear arguments (e.g., `/review-pr 123`), reducing cognitive load and execution time. Multi-purpose commands handle varied tasks but increase complexity; best practice favors single-purpose design for automation workflows.