Slash Command Generator

How to use it

1. Pick a preset, or type a command name and write the prompt you want to reuse.
2. Choose whether it takes arguments, and add any shell commands whose output Claude should see first.
3. Check the list on the right. Fix anything marked as an error.
4. Copy or download the file and save it in the folder shown under the preview.
5. In Claude Code, type the slash and the name, for example /fix-issue 123.

What a slash command file looks like

A custom slash command is a Markdown file. The file name is the command name, so commit.md becomes /commit. The optional settings between the two --- lines describe the command. Everything below is the prompt Claude receives. Project commands live in .claude/commands/ and are shared through your repo. Personal ones live in ~/.claude/commands/.

---
description: Fix a GitHub issue by number
argument-hint: "[issue-number]"
allowed-tools: Bash(git status:*), Bash(gh issue view:*)
model: sonnet
---
## Context

- Current status: !`git status`

Fix GitHub issue #$1. Start by running
gh issue view $1 and read the discussion.
Make the smallest change that fixes the cause
and add a test for it.

• description: the short text shown in the command menu.
• argument-hint: a reminder of what to type after the command. Wrap it in quotes when it starts with a square bracket.
• allowed-tools: tools the command may use without asking. Shell commands are written like Bash(git status:*).
• model: which model runs the command. Leave it out to use your current one.
• $ARGUMENTS is everything typed after the command. $1, $2 and so on are the first, second and later values.
• A line such as !`git status` runs that command first and puts its output into the prompt. It needs a matching Bash(...) entry in allowed-tools.
• @path/to/file includes that file's contents.

FAQ

What is the difference between a slash command and a subagent?

A slash command is a saved prompt that you run yourself by typing it. A subagent is a separate helper with its own instructions that Claude can hand work to. Use a command for a task you repeat, such as writing a commit message.

Do I have to add arguments?

No. A command with no arguments works fine. Add them when the same prompt needs a different input each time, like an issue number or a file path.

Why does my command ask for permission to run a shell command?

Lines that start with an exclamation mark and a backtick-wrapped command only run if the command is listed in allowed-tools. This tool adds those entries for you when you use the "Context to gather first" box.

What are subfolders for?

They let you group related commands, such as everything for git in .claude/commands/git/. The subfolder name shows next to the command in the menu.

Does this tool send my text anywhere?

No. Everything runs in your browser and nothing is uploaded or stored.

Want ready-made specialists instead of writing your own? Agentsforce ships 38 ready-made specialists for Claude Code.