Here’s the complete reference sorted by what they actually do:
◆ /init
↳ Reads your repo and writes the project brief
◆ /plan
↳ Switches to read-only mode and maps every step before touching a file
◆ /context
↳ Shows what is eating your tokens
◆ /compact
↳ Frees up context mid-session without losing the thread
◆ /clear
↳ Full reset between tasks so old context stops poisoning new output
◆ /model
↳ Swap Opus, Sonnet, Haiku mid-session
◆ /agents
↳ Spin up sub-agents and split big jobs across parallel workers
◆ /chrome
↳ Claude controls your browser. Clicks, fills, reads logs.
◆ /loop
↳ Run a prompt on a recurring schedule for up to 3 days
◆ /rewind
↳ Undo when Claude goes off track
◆ /ultraplan
↳ Forces max reasoning depth for one turn
The graphic has all 89 in one place.
Save it. You’ll come back to it.
📌 Join 1,600+ AI content creators in my free community: (Link in bio)
♻️ Repost to help someone in your network use Claude Code properly.
➕ Follow Us for more on AI and LinkedIn growth.
