Create an emoji-sticker set concept based on my dog
When to use it
You want a sticker-set concept of your own dog from an image generator and have photos or a description of the dog ready.
Make your own AI image promptWhat came out
The Midjourney stand-in kept aspect ratio and resolution from the instruction (0 of 2 defaulted) and treated the set as one generation.
Run logThis is what Mephify is programmed to produce, so it is AI-authored output, published exactly as it came out. The tabs compare how the one-line request above unfolds into each target AI's format.
## Output structure
**Main prompt**
[Main image-generation direction for the dog and sticker set]
**Negative prompt**
[Concrete visual exclusions]
**Parameters**
- Format: emoji sticker set
- Sticker count: [FILL IN: sticker count]
- Aspect ratio: [RECOMMENDED: 1:1 square — emoji stickers need compact, legible individual frames]
- Resolution: [RECOMMENDED: 2048 × 2048 px — suitable for detailed digital sticker artwork; reduce one level if unavailable]
- Background: transparent or [FILL IN: required background format]
**Continuity procedure**
[Identity and series-consistency instructions]
**Handoff note**
[One sentence explaining any directorial choice that changes or clarifies the initial request]
## Render direction
**Content for the image field — noun-phrase block**
[dog description, copied exactly from the user's supplied wording]
expressive emoji sticker character, clear silhouette, readable facial expression, one distinct action per sticker, bold outer contour, simplified shapes, clean colour blocks, controlled highlights, consistent head-to-body proportions, consistent muzzle and ear shapes, consistent coat pattern, consistent eye colour, consistent collar or accessories if supplied, centered composition, generous transparent margin, uncluttered foreground, cohesive sticker-sheet visual language, playful gesture variation, legible at small size
Use the user's actual dog description wherever the bracketed description appears; do not infer breed, colours, markings, age, accessories, personality, or body shape. If the user supplies a preferred style, use it as the governing style. Otherwise use the concrete visual treatment above rather than adding vague quality claims.
The set should vary through pose, expression, and action while returning to the same dog identity, outline treatment, scale, and colour handling. Choose each sticker's action because it communicates a distinct everyday reaction; do not add decorative details without a visual purpose.
Hybrid boundary: keep the engine-facing direction as comma-separated noun phrases. Keep decisions, assumptions, and handoff guidance as short explanatory sentences outside that block. The decisive passage is the identity-and-expression system, so give it more detail than the general stylistic description.
## Exclusions
unidentified dog traits, altered coat markings, different ear shape, different muzzle shape, inconsistent eye colour, changing collar, extra limbs, duplicated paws, distorted anatomy, cropped ears, cropped tail, unreadable tiny details, cluttered background, realistic photographic background, text, captions, speech bubbles, watermark, logo, signature, frame border, muddy colours, over-rendered fur, harsh glare, frightening expression, accidental aggression, duplicate stickers in one frame
## Continuity procedure
The supplied dog description is the fixed identity anchor. Copy its wording character-for-character into every sticker prompt. Identity must remain stable through the same coat pattern, facial proportions, eye colour, ear shape, muzzle shape, outline weight, colour palette, and accessory treatment.
On the first generation, settle any unnamed visual facts in one sentence: use a consistent simplified background treatment, transparent negative space around the silhouette, and the same front-facing three-quarter character orientation unless an action requires a readable side view. Record those values and quote the sentence unchanged in later generations.
If the service supports a fixed seed, record the first image's seed and reuse it. If it supports a reference image or character-reference input, feed the first approved image into later generations. Wording alone will not reliably reproduce the same dog: reference input is stronger than a fixed seed, and a fixed seed is stronger than text description. A seed remains useful only while the prompt stays unchanged.
## Parameters and handoff
- Dog identity: [FILL IN: dog description — provide coat, markings, face, ears, eyes, body shape, and any fixed accessory]
- Sticker count: [FILL IN: sticker count — provide the number of stickers required]
- Intended use: [FILL IN: platform or use — provide the destination if it imposes file, transparency, or size rules]
- Visual style: [FILL IN: preferred style, if any — name the style or describe its line, colour, and rendering treatment]
- Aspect ratio: [RECOMMENDED: 1:1 square — keeps each expression readable as an individual emoji]
- Resolution: [RECOMMENDED: 2048 × 2048 px — preserves clean edges during resizing; the service may not offer this exact size]
- Background: [RECOMMENDED: transparent — allows placement across chat backgrounds; change this if the destination requires a solid background]
- Seed policy: record and reuse the first approved seed if supported; otherwise use the first approved image as a reference if supported
- Output: one separate image per sticker, consistent margins, no text unless the user explicitly supplies text
I will revise after you choose one path:
1. Fill in the dog details, sticker count, use, and style, then I will complete the set.
2. Finish it with the recommended values and a neutral dog-description slot.
3. Change the sticker-set structure first.
1. **[FILL IN: dog description]** — supply the visual traits that must remain recognizable.
2. **[FILL IN: sticker count and intended use]** — supply the number of stickers and where they will be used.
3. **[FILL IN: preferred visual style, if any]** — supply a named style or concrete line-and-colour preferences.6 things this instruction asks you for
Collected from the [FILL IN] slots inside the instruction. Have these ready before pasting and the target AI will not have to ask; leave what you don't know empty and it stays a slot rather than a guess.
- sticker count
- required background format
- sticker count — provide the number of stickers required
- dog description
- sticker count and intended use
- preferred visual style, if any
Operator's verification log
We actually ran it
2026-09-18 · The generation engine cannot be run here, so the instruction above was given to a stand-in that plays that engine and reports the decisions it would make (one run each). A stand-in is not the real thing — it finds holes in the instruction; it does not judge the output. The instruction itself was not edited.
- Midjourney version → Midjourney stand-in (ChatGPT) · 1 units · 10 guesses · 3 inert sentences · 0/2 params defaulted
- What you must fill in
- Your dog's breed, coat colour and markings, ear shape, muzzle, eye colour and build; how many stickers; the emotion or action for each; background (transparent or flat); outline and palette.
- What came out (continued)
- It guessed 10 things — everything about the dog itself, the sticker count, each sticker's emotion, the background, outline thickness, palette, eye style — because the request never described the dog.
- What we flagged
- Without a fixed description of the dog the identity anchor cannot work: the stand-in reported it would keep outline, palette and eye treatment consistent but could not guarantee the same dog twice. Describe the dog in the slots, and if your service supports a reference image or a fixed seed, use it — the instruction says so, conditionally.
- Where not to use it
- A stand-in check, not generated images. It cannot look at your photos; you have to describe the dog in words.