Good AI image prompts name the subject first, then add what it looks like, what it is doing, the setting, the light, the mood and the framing, in one or two plain sentences. Pick the art style separately when your generator has style options, and change one detail at a time until the picture looks right.
This guide gives you a simple prompt formula, 12 ready AI image prompt examples sorted by use, fixes for the most common bad results, and what you may do with the images you make.
What makes a good AI image prompt?
A good AI image prompt is specific, visual and short. It says what should be in the picture, where it is, how it is lit and how close the camera is, using words for things you could actually see. "A dog" leaves every choice to the AI; "a golden retriever puppy asleep on a red blanket, soft window light, close-up" gets you much closer to what you pictured.
Three things make the biggest difference:
- Concrete nouns and adjectives: "a brass desk lamp" rather than "a nice lamp".
- Light and mood: "golden hour", "foggy morning", "neon at night".
- One clear style, not three that fight each other.
The AI image prompt formula
Use this order: subject + details + action + setting + light + mood + framing. Not every prompt needs every part, but the subject always comes first so it is not lost among the details.
| Part | What to write | Example |
|---|---|---|
| Subject | The main thing in the picture | an old fisherman |
| Details | Colors, materials, clothes, age | grey beard, yellow raincoat |
| Action | What it is doing | mending a net |
| Setting | Where and when | on a wooden pier, early morning |
| Light | Type and direction of light | soft light from behind, light fog |
| Mood | The feeling | calm, quiet |
| Framing | Camera distance and angle | medium shot, eye level |
Put together: An old fisherman with a grey beard in a yellow raincoat, mending a net on a wooden pier in the early morning, soft light from behind, light fog, calm and quiet, medium shot at eye level.
That is under 200 characters. In the free WoowTools AI image generator a prompt can be up to 500 characters, and English prompts work best.
How to write AI image prompts step by step
- Write the subject in a few words: "a lighthouse on a rocky cliff".
- Add two or three details you care about: colors, materials, time of day.
- Describe the light. It changes the result more than almost any other word.
- Choose the framing: close-up, wide shot, top-down view, eye level.
- Pick the style and the shape in the tool, not in the text (see the next section).
- Generate, look, change one thing. If only the sky is wrong, change only the sky words and run it again.
Should you write the style in the prompt?
Usually not. Our generator has 9 style buttons, and each one adds its own style wording to your prompt behind the scenes. Photo, for example, asks for a realistic, naturally lit photograph and steers away from paintings and cartoons. So describe the scene in your prompt and let the button set the look.
| Style button | Best for |
|---|---|
| No style | Full control: it only asks for fine detail and sharp focus |
| Photo | Realistic scenes, products, food, nature |
| Cinematic | Dramatic covers and thumbnails |
| Anime | Characters and bright illustrations |
| Digital art | Concept art, fantasy and game scenes |
| 3D render | Icons, toys, product mockups |
| Watercolor | Cards, prints, soft illustrations |
| Oil painting | Classic, painterly portraits and landscapes |
| Pixel art | Retro game looks and avatars |
Avoid mixing: writing "watercolor painting" in your prompt while the Photo button is on asks for two opposite looks. Choose No style when you want to describe the style entirely in your own words.
Pick the shape the same way: Square (1024 x 1024 pixels) for posts, Wide 16:9 (1344 x 768) for banners and thumbnails, Tall 9:16 (768 x 1344) for stories and phone screens, or 4:3 and 3:4 for slides and prints.
AI image prompt examples by use
Each prompt below is ready to paste. The style and shape columns are the buttons to pick.
| Use | Prompt | Style | Shape |
|---|---|---|---|
| Blog cover | A tidy home office desk with a laptop, a green plant and a cup of tea, morning light through a window, calm mood, wide shot | Photo | Wide 16:9 |
| Product shot | A glass perfume bottle on a smooth stone, soft shadows, beige background, close-up | Photo | Square |
| Food | A bowl of ramen with a soft-boiled egg and spring onions, steam rising, dark wooden table, top-down view | Photo | Square |
| Video thumbnail | A lone hiker on a mountain ridge at sunrise, dramatic clouds, orange sky, wide shot | Cinematic | Wide 16:9 |
| Phone wallpaper | A quiet lake surrounded by pine trees at night, stars reflected in the water | Digital art | Tall 9:16 |
| Story background | Pastel pink and blue clouds at sunset, soft and dreamy | Watercolor | Tall 9:16 |
| Character | A wizard with round glasses and a blue cloak holding a glowing book, old library in the background | Anime | 3:4 |
| App icon idea | A cute rocket made of smooth plastic, rounded shapes, bright orange and white, plain background | 3D render | Square |
| Greeting card | A bunch of wildflowers in a glass jar on a windowsill, spring morning | Watercolor | 3:4 |
| Wall art | A harbor town at dusk with fishing boats and warm lights in the houses | Oil painting | 4:3 |
| Game asset | A small wooden treasure chest full of gold coins on green grass | Pixel art | Square |
| Presentation slide | An aerial view of a winding river through green farmland, soft afternoon light | Photo | 4:3 |
What to do when the result is off
When an image comes out wrong, change one part of the prompt at a time and generate again. Every run gives a different picture, even with the same words, so a second or third try often fixes small flaws.
| Problem | Fix |
|---|---|
| The subject is small or missing | Move it to the start and add "close-up" or "medium shot" |
| Too busy or cluttered | Remove extra objects; keep one subject and one setting |
| Wrong look or mixed styles | Pick one style button and remove style words from the prompt |
| Odd hands or faces | Frame from a little farther away or give the hands a simple pose, then generate again |
| Garbled letters or signs | Leave text out and add it afterwards |
| Wrong colors | Name the colors of the subject and of the background |
| Flat or dull | Add a light type: golden hour, rim light, soft window light |
Do you need negative prompts?
Not with our generator. It has no separate negative prompt box because it already adds a built-in list of things to avoid to every image, including blur, low quality, watermarks, text, logos and extra fingers. Also avoid writing "no people" or "without trees" in the main prompt: image models can latch onto the noun and add exactly what you wanted to leave out. Describe what you do want instead, such as "an empty beach at dawn".
Finish the image before you publish
- Add a title or caption with add text to image, since AI images rarely spell words correctly.
- Cut out the subject with the background remover for product shots and stickers.
- Make it bigger for print or large screens with the image upscaler.
Can you use AI generated images?
You may use the images you create with the WoowTools generator in personal and commercial projects, such as blogs, social posts, ads and presentations, and there is no watermark. Sexual content, anything that sexualizes minors, graphic violence and images meant to deceive or harass real people are not allowed, and a safety filter blocks the obvious cases.
Owning the image is a separate question. The U.S. Copyright Office concluded in January 2025 that prompts alone do not make you the author of an AI output; protection can come from human expression you add, such as your own edits or creative arrangement. If copyright matters for your project, add your own work to the image and check the rules where you live.
Try your first prompt
Open the free AI image generator, paste one of the examples above, pick the style and shape from the table and click Generate image. It usually takes under a minute, you download the result as a PNG. The free daily limit is about 15 images per visitor, with no sign-up.
Frequently asked questions
What is a good prompt for AI art?
A good prompt names the subject first, then adds details, action, setting, light, mood and framing. For example: an old fisherman in a yellow raincoat mending a net on a wooden pier, soft morning light, medium shot. Pick the art style with the tool's style option.
How long should an AI image prompt be?
One or two sentences is usually enough, about 100 to 300 characters. Our generator accepts up to 500 characters. Longer prompts with many ideas tend to give cluttered results.
Why does my AI image have weird hands or faces?
Hands and small faces are hard for image models. Frame the subject from a little farther away, give the hands a simple pose, and generate again, since every run is different.
Can an AI image generator write text in the picture?
Not reliably. AI images often show garbled letters, and our generator steers away from text by default. Add titles afterwards with a text on image tool.
Will the same prompt give the same image again?
No. Each run starts from a new random point, so the same prompt gives a different image every time. Save the one you like by downloading the PNG.
Is the WoowTools AI image generator free?
Yes. It is free with a daily limit of about 15 images per visitor, with no sign-up and no watermark. Images usually take under a minute.
- ai image generator
- ai prompts
- text to image
- ai art
Share
