19d • General Discussion 💬
Build your own slash commands | Claude Code
Easy peasy - most people don't know you can create custom /commands inside Claude Code. This can drastically improve speed and consistency, without having to type it out every. single. time.
Here's the full setup:
Step 1 — Create the folder
- From your project root, run: mkdir -p .claude/commands
(Want it available across all your projects? Use ~/.claude/commands/ instead)
Step 2 — Create a .md file in that folder
- The filename = the command name. Example: update-weather.md
Step 3 -- Write the file
Pick and mix from the three methods below.
- Method 1: Static prompt command (e.g., "review.md")
Review the current status file for:
- Logic errors or edge cases
- Naming clarity
Give me a prioritized list, most critical first.
Usage: /review → fires that exact prompt instantly.
- Method 2: Command with arguments (e.g., "commit.md")
Write a git commit message for: $ARGUMENTS
Follow conventional commits format. Be specific, not generic.
Usage: /commit added retry logic to the webhook handler
The $ARGUMENTS placeholder gets replaced with whatever you type after the command.
- Type 3: Context injection (no question needed, e.g., "scope.md")
Before we start, here is the architecture context for this project:
- Stack: Next.js + Supabase + GHL webhooks
- Auth is handled by middleware, never in components
Acknowledge this and wait for my next instruction.
Usage: /scope at the start of any session to prime Claude before a big task.
That's it. No config changes. I'd recommend just opening a new Claude Code thread, but you could just drop the file and the command is live.
6
6 comments
Build your own slash commands | Claude Code
skool.com/ai-automation-society
Learn to get paid for AI solutions, regardless of your background.
Leaderboard (30-day)
1
+5431
2
🔥
+3903
3
+3243
4
+2142
5
+1153
Powered by