Back to All Cheatsheet Libraries cheatsheets

ChatGPT

Shortcuts and key features (Memory, Projects, Custom GPTs), prompt-structure patterns, and habits for getting better answers faster.

Total Items: 0
Category Action macOS Windows
ChatNew chat⇧⌘OCtrl+Shift+O
ChatFocus chat input⇧EscShift+Esc
NavigateSearch chat history⌘KCtrl+K
InputToggle voice input⇧⌘UCtrl+Shift+U

Key Features

Custom Instructions

Persistent context about you and how you want responses shaped — applied automatically to every new conversation without re-explaining each time.

Memory

Lets the model retain specific facts across separate chats when enabled — distinct from Custom Instructions, and viewable/editable/deletable in settings.

Projects

Groups related chats and files together with shared instructions — useful for keeping a long-running task's context in one place.

Code Interpreter / Data Analysis

Runs actual Python in a sandbox — for real calculations, file parsing, and chart generation rather than the model just describing what it would do.

Custom GPTs

A saved configuration of instructions, knowledge files, and enabled tools — packaged as a reusable, shareable assistant for a specific task.

Prompt Structure Patterns

Role

"Act as a senior..."

Task

The specific, concrete ask

Context

Background the model needs

Format

"as a table", "under 200 words"

A well-formed prompt example

Combining role, context, task, and format into one clear request.

You are an experienced technical editor. Context: The following paragraph is from a beginner's guide to Docker. Task: Simplify it for someone who has never used the command line. Format: Return only the rewritten paragraph, no preamble. [paste paragraph here]

Iteration Techniques

Technique When to Use
Few-shot examplesShow 2-3 examples of the exact input/output pattern wanted before the real request.
Chain-of-thought ("think step by step")For reasoning/math-heavy tasks — asking for stepwise reasoning improves accuracy on multi-step problems.
Ask it to critique its own answerA follow-up "what's wrong with this?" often surfaces errors the first pass missed.
Constrain the output format explicitly"Return only valid JSON, no markdown" — prevents extra commentary from breaking a programmatic parse.

Getting Better Answers Faster

The habits that consistently cut down on back-and-forth clarification.

1

Front-load context, don't drip-feed it

Paste the relevant code/document/background in the first message rather than adding it piece by piece across several replies.

2

State the output format you want

"As a table", "as a bulleted list", "under 100 words" — removes an entire category of unhelpful-formatting follow-ups.

3

Split a big ask into stages

"First outline, then I'll ask you to write" beats one giant prompt for anything with real structure — each stage is checkable before moving on.

4

Set Custom Instructions once

Your role, preferred tone, and common constraints saved once apply to every future chat — far less repetitive than restating them each session.

Quick Tips

Verify facts, especially numbers and citations
Models can state incorrect information confidently — treat specific figures, quotes, and citations as claims to verify, not settled facts.
Don't paste real secrets into a prompt
API keys, passwords, and other credentials shouldn't go into any chat, regardless of the account's data-retention settings.
A bad answer often means an ambiguous prompt
Before assuming a limitation, try rephrasing with the specific constraint or example that was actually implied but not stated.