GUIDES
The Claude Code Commands I Use Every Day 📚
You commented SLASH, so here it is. This is the full list of my favorite Claude Code commands, all 13 of them, grouped by when to reach for each one. Save this.
You commented SLASH, so here it is. All 13 of my favorite Claude Code commands in one place, grouped by when you actually reach for them, each with what it does and the exact syntax to type. Save this page and come back to it.
Everyday Essentials
Edit your CLAUDE.md files so Claude remembers your facts, preferences, and rules across future sessions, not just this one.
Switch which AI model you are using and save it as your default. Jump up for hard tasks like a big refactor, drop down for quick edits.
Roll the conversation and code back to an earlier checkpoint. When Claude goes down a wrong path, do not argue with it, just rewind to a clean point and start fresh.
Power Moves
Puts Claude in plan mode: it writes a full step-by-step plan and changes nothing until you approve it. Use it before anything big or risky.
Set a goal condition and Claude keeps working on its own until it is met, like all tests passing. Set it loose and walk away.
Create specialized subagents with their own instructions and delegate tasks to them. Great for big jobs or anything you do over and over.
Detaches the session to run in the background so it keeps working while your terminal stays free. Great for long jobs you do not want to babysit.
Context & Cost Control
Shows a live breakdown of how full your context window is, so you can see what is filling it up before Claude starts forgetting things.
Summarizes the conversation so far to free up context without starting a new chat. Add a short note to control what survives.
Shows your session cost, plan usage limits, and activity stats all in one place, so you always know where you stand.
Safety & Automation
Copies the current conversation so you can try a risky idea without losing your original path. Do not like it? Your original is right where you left it.
Scans your pending changes for security problems like injection, auth issues, and data exposure before they go live. Run it right before you ship.
Runs a prompt over and over while your session stays open, so it can keep watch or run a check on repeat while you do other things.
My Daily Flow
Here is how they chain in a real session: I start with /plan, set a /goal and walk away, watch with /context and trim with /compact, /rewind if it wanders, then run /security-review before I ship and check /usage to see where I landed. Save this page and go build something.
The real unlock of AI agents is not full autonomy, but human-in-the-loop systems design. Structure your agent inputs with explicit constraints, negative prompts, and automated test checkpoints. When building tools, keep token consumption lean by caching system prompts and isolating tasks into specialized sub-agents.