Emoticon and Sticker Set Generator

Create a consistent sticker pack featuring various emotions and poses.

Prompt · 2 variables

I want to create a cohesive sticker pack for messaging apps. The character is "{{character}}". I frequently run into the issue where each image looks fine individually, but together they look like completely different characters. I want to prevent this inconsistency.

First, please create a standardized, single-line anchor description in English for this character. It must include body proportions, facial features, color palette, signature items, and art style. Next, generate a single-line image generation prompt for each item listed in {{emotion_list}}. Every line must start with the exact, unaltered anchor description, followed only by the specific expression and action for that emotion.

Here is an example format of a single output prompt. Do not copy the wording directly; use it only as a reference for length and structure. Example: a chubby round penguin with a red scarf, short stubby wings, thick clean outline, flat pastel colors, raising both wings high with sparkling eyes, plain white background, full body sticker

Provide the output in a table format. The left column should be the emotion/situation, and the right column should be the single-line English image prompt. Place the standardized anchor description on a separate line above the table.

Do not include any text, letters, or words inside the generated images, as AI-generated text often renders poorly and captions are better added separately. Ensure the character is always shown in full-body view with ample padding from the canvas edges. Keep the background plain solid white, and do not replicate or closely mimic existing commercial characters or copyrighted mascots.

Copy, then paste here · ChatGPT and Claude open with the prompt filled in Open in ChatGPT ↗Open in Claude ↗Open in Gemini ↗ Edit in builder Download classroom card

Why it is written this way

Context
I want to create a cohesive sticker pack for messaging apps. The character is "{{character}}". I frequently run into the issue where each image looks fine individually, but together they look like completely different characters. I want to prevent this inconsistency.
Task
First, please create a standardized, single-line anchor description in English for this character. It must include body proportions, facial features, color palette, signature items, and art style. Next, generate a single-line image generation prompt for each item listed in {{emotion_list}}. Every line must start with the exact, unaltered anchor description, followed only by the specific expression and action for that emotion.
Example
Here is an example format of a single output prompt. Do not copy the wording directly; use it only as a reference for length and structure. Example: a chubby round penguin with a red scarf, short stubby wings, thick clean outline, flat pastel colors, raising both wings high with sparkling eyes, plain white background, full body sticker
Format
Provide the output in a table format. The left column should be the emotion/situation, and the right column should be the single-line English image prompt. Place the standardized anchor description on a separate line above the table.
Constraints
Do not include any text, letters, or words inside the generated images, as AI-generated text often renders poorly and captions are better added separately. Ensure the character is always shown in full-body view with ample padding from the canvas edges. Keep the background plain solid white, and do not replicate or closely mimic existing commercial characters or copyrighted mascots.

The hardest part of generating an emoticon or sticker pack with AI is maintaining consistency across the entire set rather than a single image. If you ask for eight different emotions at once, the AI reimagines the character every time—dropping glasses, changing body proportions, or altering colors. Because stickers appear side-by-side in chat windows, these discrepancies are immediately noticeable.

The Task structure breaks the workflow into two clear steps: lock down an anchor description first, then swap out only the expression and pose. Forcing body proportions and signature items into the anchor description prevents the two most common tells of character drift.

The Example replaces several paragraphs of formatting instructions. It demonstrates the ideal phrase length, comma placement, and slot positioning for actions. Using a different animal in the example and explicitly stating "do not copy the wording" stops the model from regurgitating example keywords.

The Format specifies a table so you can instantly verify that every requested emotion has an exact 1:1 prompt pairing. The Constraints regarding full-body framing, border margins, and clean white backgrounds are practical essentials for sticker creation: cut-off edges or messy backgrounds make asset extraction difficult and look awkward in chat interfaces. Keeping the subject focused on one clear action against empty space ensures readability at small icon sizes.

Unfamiliar terms? See Aha AI: few-shot, output-format

Compared with a bad example

Common bad example

Make prompts for emoticons of a gray cat with glasses for thumbs up, thanks, sorry, leaving work, and sleeping.

A prompt like this causes the cat's description to vary on every line. Some lines will omit the scarf, others will stretch the torso into human proportions, and lack of background/framing constraints will yield mixed half-body crops and colored backdrops. Ultimately, only two or three images out of the set will be usable.

Variations

Adding New Emotions to an Existing Set

Adding New Emotions to an Existing Set

I want to add a few more emotions to the previously created sticker set. Keeping the existing anchor description intact, please generate a table in the exact same format only for the new items listed in {{emotion_list}}. Check to ensure the expressions and poses do not overlap with the previous ones, and point out any duplicates if they exist.

When expanding a sticker pack, this variant preserves character consistency while preventing duplicate poses.

Creating Face-Only / Close-Up Stickers

Creating Face-Only / Close-Up Stickers

I want to create close-up face stickers for "{{character}}". Please revise the anchor description by removing body-related details and keeping only the facial features, signature accessories around the face, color palette, and art style. Then, write prompts focusing strictly on bold, unmistakable facial expressions. Specify large, simplified facial cues so they remain readable even at tiny thumbnail sizes.

Face-focused stickers require different visual density than full-body poses. This variant trims the anchor description to maximize facial readability on small screens.

Model notes

Some AI image generators produce transparent PNGs natively, while others default to solid white backgrounds. For the latter, you may need a background removal tool as a final post-processing step.

Related prompts

Last updated 2026-09-02 · Found a mistake? Let us know