ChatGPT Cheat Codes 2026: 500+ Prompt Shortcuts, Image Codes & AI Commands

ChatGPT is no longer just a chatbot. People use it to learn, write, research, analyse data, build presentations, generate images, edit photographs, plan marketing campaigns, write code and design entire workflows. Along the way, thousands of so-called "ChatGPT cheat codes" started circulating on Instagram, YouTube, LinkedIn and Reddit - /eli5, /humanize, /mindmap, /poster, /cinematic, /productphoto and hundreds more. This library brings them together in one place, honestly labelled.

Official commands vs prompt shortcuts

Official ChatGPT functionality

ChatGPT has genuine tools and capabilities: image generation, image editing, image analysis, web search, deep research, file analysis, data analysis, coding, Projects, Custom GPTs, voice, and connectors that depend on your account. OpenAI documents ChatGPT Images as a real feature that can create and edit images, accept uploads, change aspect ratios, add or remove elements and work through iterative edits.

Prompt shortcuts

Reusable phrases such as /eli5, /critique, /summarize, /mindmap, /humanize and /compare. Shorthand for instructions, nothing more.

Tool-specific syntax

Some platforms have genuine parameters. Midjourney, for example, has official parameters like --ar, --chaos, --no, --quality, --seed, --stylize, --sref, --tile, --weird and --version. These are different from generic ChatGPT prompt shortcuts and belong in their own section.

Part 1: Learning cheat codes

/eli5

Explain blockchain to a 12-year-old using:
1. One analogy
2. One real-life example
3. One simple diagram
4. Five key points

Part 2: Writing cheat codes

Part 3: Business cheat codes

Part 4: Sales cheat codes

/objection

Customer says: "Your insurance premium is too expensive."

Create:
1. Empathetic response
2. Discovery question
3. Value-based response
4. Proof point
5. Closing question

Do not sound pushy.

Part 5: Marketing cheat codes

Part 6: SEO, AEO and GEO cheat codes

/aeo

Topic: AI training in India

Create:
1. 10 questions people ask AI search engines
2. Direct answers
3. Supporting entities
4. FAQs
5. Evidence requirements
6. Suggested schema
7. Internal links

Part 7: Social media cheat codes

Part 8: Video creation cheat codes

Part 9: Image cheat codes

ChatGPT Images supports both creation and editing, including editing uploaded images, adding or removing elements, aspect-ratio changes and transparent backgrounds. The codes below are visual vocabulary that makes your intent unambiguous.

Google's current Gemini image guidance similarly recommends explicitly describing subject, setting, action, composition and style, then refining through iterative edits.

Product photography

A strong product prompt specifies the product, camera view, lighting, surface, background, composition and what must remain unchanged. Reference images matter most when the exact product appearance is non-negotiable.

/productphoto

Use my uploaded product image as the exact product reference.
Create a premium commercial product photograph.

Product: [PRODUCT]
Environment: [ENVIRONMENT]
Camera: [ANGLE]
Lighting: [LIGHTING]
Background: [BACKGROUND]
Surface: [SURFACE]
Composition: [COMPOSITION]

Keep the product's shape, logo, colors, proportions,
packaging and text unchanged.

For consistency across a series, use a clear reference image and name the recurring subject explicitly - Google's image guidance recommends assigning distinct names to recurring subjects.

Text inside images

Instead of "make a poster about AI", write:

Create a professional event poster.

Headline: "AI MASTERCLASS 2026"
Subheadline: "Learn ChatGPT, Gemini, Claude & AI Automation"
Style: Premium corporate technology
Layout: Headline at top, speaker area in centre,
        date and venue at bottom

Leave sufficient negative space.
Do not add any additional words.

These are creative shortcuts, not official ChatGPT commands. Similar lists circulate in creator communities, particularly for YouTube thumbnail art direction.

ChatGPT Images supports selecting an area and describing an edit directly, although edits can sometimes extend beyond the selected region - always review the result.

Part 10: Work, code and analysis cheat codes

ChatGPT's Deep Research capability is built for multi-step research across online sources with structured synthesis, so treat research workflows as different from simple answer-generation shortcuts.

The master ChatGPT cheat code

Instead of memorising hundreds of commands, use this structure:

/MASTER

ROLE: Act as [ROLE].
TASK: [WHAT YOU WANT]
CONTEXT: [BACKGROUND INFORMATION]
AUDIENCE: [WHO WILL USE IT]
CONSTRAINTS: [WHAT TO AVOID]
OUTPUT: [DESIRED FORMAT]
TONE: [TONE]
QUALITY: [WHAT GOOD LOOKS LIKE]
EXAMPLES: [OPTIONAL EXAMPLE]

CHECK: Before finalizing, review the answer for accuracy,
completeness and alignment with the instructions.

The master image cheat code

/IMAGE MASTER

Create a [IMAGE TYPE] of [SUBJECT].

PURPOSE: [ADVERTISEMENT / WEBSITE / INSTAGRAM / PRODUCT / DECK]
SUBJECT: Describe the main subject precisely.
ACTION: What is happening?
ENVIRONMENT: Where is the scene?
STYLE: [PHOTOREALISTIC / CINEMATIC / ILLUSTRATION / 3D]
COMPOSITION: [PORTRAIT / CLOSE-UP / WIDE / TOP-DOWN]
CAMERA: [CAMERA FEEL / LENS / ANGLE]
LIGHTING: [LIGHT SOURCE / DIRECTION / QUALITY]
COLORS: [COLOR PALETTE]
MOOD: [MOOD]
BACKGROUND: [BACKGROUND]
TEXT: "[EXACT TEXT]"
ASPECT RATIO: [1:1 / 4:5 / 16:9 / 9:16]
REFERENCE: Use uploaded image(s) for [IDENTITY / PRODUCT / STYLE]
IMPORTANT: Preserve [ELEMENTS].
AVOID: [UNWANTED ELEMENTS].

Create 3 distinct creative directions.

This formula matches the image-prompt guidance published by OpenAI, Google and Adobe: specify subject, environment, action, style, composition, lighting, colour and output format, then iterate.

10 powerful image combinations

The real power is not one code. It is stacking codes.

AI image tool cheat sheet

Not every image tool uses the same syntax.

Midjourney parameters (genuine syntax)

These are actual Midjourney parameters, not ChatGPT shortcuts. Midjourney also supports image prompts, multiple image prompts, image-weight control and style references.

Gemini image cheat sheet

Gemini's official guidance recommends describing Subject → Setting → Action → Composition → Style, then iterating through conversational edits. Useful shorthand labels (ours, not official commands):

Adobe Firefly formula

[IMAGE TYPE] + [SUBJECT] + [ACTION] + [COMPOSITION]
+ [LIGHTING] + [COLOUR] + [MOOD] + [BACKGROUND]

Adobe recommends specificity around style, composition, perspective, lighting, atmosphere, colour, mood, style-specific terminology and iteration.

The biggest cheat-code myth

You will see posts claiming "ChatGPT has 100 secret image commands". Be careful. Community creators do publish lists like /poster, /cottagecore, /blueprint, /productphoto and /cinematic, but these are prompt vocabulary or community-created shorthand, not hidden functionality. Remove the slash and the descriptive instruction still works. /cinematic portrait and "Create a cinematic portrait" differ only in presentation.

Starter sets: the codes worth memorising

The real secret: stack your cheat codes

Treat codes as building blocks, not magic commands.

/productphoto + /luxury + /studiophoto + /blackbackground
+ /rimlight + /85mm + /closeup

/research + /competitor + /swot + /market + /actionplan

Cheat code builder

/cheatcode

My objective: [DESCRIBE WHAT I WANT]

Suggest:
1. The best cheat code
2. Three supporting cheat codes
3. The final combined prompt
4. The expected output format
5. One improved version

Cheat code iteration system

Do not regenerate from scratch. Change one variable at a time.

/iterate

Keep everything unchanged.
Only modify: lighting.
Change lighting from studio to golden hour.

Do not change: subject, composition, wardrobe,
background, camera angle, colour palette.

The Nirmal Rabari 7-layer prompt formula

1. Role - who should the AI act as?
2. Objective - what should it accomplish?
3. Context - what information does it need?
4. Constraints - what must it avoid?
5. Output - what should the result look like?
6. Quality - what makes the answer good?
7. Iteration - how should the result be improved?

For images, add a visual layer: Subject → Environment → Action → Composition → Camera → Lighting → Style → Colour → Mood → Format.

Frequently asked questions

Are ChatGPT cheat codes official commands?

Not necessarily. Many commonly shared codes are community-created prompt shortcuts. ChatGPT's actual features and tool commands are separate from these shortcuts.

Does /eli5 unlock a hidden ChatGPT mode?

No. It is shorthand for asking ChatGPT to explain something very simply.

Does /image work?

ChatGPT has a genuine image-generation capability, but /image is not a secret unlock code. You can simply ask ChatGPT to create an image or use its Images interface.

Can ChatGPT edit existing photos?

Yes. Upload an image and describe the changes, or select an area and request an edit.

Can ChatGPT create images with text?

Yes. ChatGPT Images can add text and details inside generated images, though complex layouts should still be reviewed carefully.

Can I use these image codes in Gemini?

Many descriptive visual keywords adapt well to Gemini because they express ordinary visual concepts, but Gemini's product features and syntax are not identical to ChatGPT's.

Can I use these codes in Midjourney?

Many visual concepts transfer, but Midjourney has its own official parameter system such as --ar, --stylize, --chaos, --no and --seed.

What is the best image prompt formula?

Subject + Action + Environment + Composition + Style + Lighting + Colour + Mood + Format + Constraints. This is broadly consistent with current prompting guidance from Google and Adobe.

Final cheat-code rule

Do not memorise 500 commands. Tell the AI what you want, give it context, define the output, specify the constraints and iterate. Cheat codes simply make that process faster.

Next steps

• Compare with the original 200+ cheat codes library.
• Learn the frameworks in the prompt engineering frameworks guide.
• Book prompt engineering training for your team.
• Follow the AI Learning Roadmap from beginner to advanced.

About this library

This library is maintained by Nirmal Rabari, AI trainer, corporate AI consultant and founder of NMR Infotech, with a focus on practical AI adoption, prompt engineering, generative AI and AI productivity. Treat it as a living resource: AI products, models, interfaces and tool-specific parameters change frequently.