If you find yourself typing the same long instructions over and over — a Wwise audit, a render pipeline, a UCS rename routine — slash commands let you save that prompt once and trigger it withDocumentation Index
Fetch the complete documentation index at: https://reaperagent.com/docs/llms.txt
Use this file to discover all available pages before exploring further.
/.
Creating a Command
- Open Settings → Commands
- Click New Command
- Give the command a short name (this becomes the trigger, e.g.
wwise-audit) - Write the prompt body in plain English — anything you’d normally type into chat
- Save
Triggering a Command
Type/ anywhere in the chat input to open the command popup. Typing additional letters fuzzy-filters the list.
- Press Enter or Tab to insert the matching command
- Press Esc to dismiss the popup
- Type
/on its own to browse all your commands
/word token appears in blue in the input. You can mix slash commands with regular text and other commands in the same message:
What Happens When You Send
When you send the message, every blue/word token is replaced with the body of its markdown file before the message reaches the agent. The agent sees the fully expanded prompt — there’s no separate “command” mode or routing.
This means:
- Commands are pure text — they don’t change which subagent runs or what tools are available
- You can write any natural-language instruction inside a command, including references to subagents (“Use the Wwise subagent to…”)
Editing and Deleting
From Settings → Commands:- Click any command to edit its name or body
- Click Delete to remove it
- Click Open in Finder to manage commands directly on disk