Claude Code Cheatsheet

1. Update Claude Code regularly

One the most important you should do regularly is to update to the latest version. Updates and bug fixes come out nearly every single day.

Terminal window
pnpm up -g --latest
# or
claude update

2. Read claude --help to familiarize yourself with the options

Some of the commands I use regularily:

Terminal window
# Resume a previous session
claude -c/--continue
claude -r/--resume

Inside Claude Code I mostly use

Terminal window
/clear  # start a new session with new context regularily
/mcp  # check the status of the MCP servers
/context  # check how much context space I have left and when it is time to start a new session

3. Keep your prompts simple

Example:

commit everything in logical chunks

Source: https://steipete.me/posts/2025/claude-code-is-my-computer

4. Use commands

One of the best commands I have encountered is this1:

Draft a detailed, step-by-step blueprint for building this project.
Then, once you have a solid plan, break it down into small, iterative chunks that build on each other.
Look at these chunks and then go another round to break it into small steps.
review the results and make sure that the steps are small enough to be implemented safely, but big enough to move the project forward.
Iterate until you feel that the steps are right sized for this project.


From here you should have the foundation to provide a series of prompts for a code-generation LLM that will implement each step.
Prioritize best practices, and incremental progress, ensuring no big jumps in complexity at any stage.
Make sure that each prompt builds on the previous prompts, and ends with wiring things together.
There should be no hanging or orphaned code that isn't integrated into a previous step.


Make sure and separate each prompt section. Use markdown. Each prompt should be tagged as text using code tags.
The goal is to output prompts, but context, etc is important as well.


Store the plan in plan.md. Also create a todo.md to keep state.


The spec is in the file called: $ARGUMENTS

Note the $ARGUMENTS keyword. It helps you pass different keywords to your command, as if you were using a command line tool.

5. Use skills

6. Use Git Worktrees (or at least branches)

Technically, you could make a directory .trees/ inside your repo and create git worktrees there2:

Terminal window
mkdir .trees
git worktree add .trees/ui_feature
git worktree add .trees/testing_feature
git worktree add .trees/quality_feature

Then, after completion:

use the git merge command to merge in all of the worktrees in the .trees folder and fix any conflicts if there are any

But I found it to conflict with code searches, since claude will find duplicate files an use up unnecessary tokens. So it is preferable to:

Terminal window
git worktree add ../ui_feature
git worktree add ../testing_feature
git worktree add ../quality_feature

7. Use MCP servers

Nice tools around Claude Code

Undo in Claude Code without wasting tokens

Use ccundo.

Terminal window
pnpm install -g ccundo


# 1. Check what changed
ccundo list


# 2. Preview the changes
ccundo preview


# 3. Undo it
ccundo undo

Read How to Undo in Claude Code? - DEV Community - it’s where I got

Check your token usage

Find out how much you consume with

Terminal window
npx ccusage@latest

Footnotes

  1. Source: My LLM codegen workflow atm | Harper Reed’s Blog ↩

  2. As described in Claude Code: A Highly Agentic Coding Assistant - DeepLearning.AI ↩