Chibi Mini-Me example output generated by Doodle AI — A real run of this skill. The input is a synthetic portrait, not a customer photo.

The photo used as input for Chibi Mini-Me — A real run of this skill. The input is a synthetic portrait, not a customer photo.

Ready to use

Chibi Mini-Me

Tiny cartoon versions of you doodled over your real photo

Use this skill

Your photo: Required
Output: Single square 1:1
Category: avatars
Cost: 1 credit

chibi
overlay
selfie
annotation

What it does

What you get back

Preserves the original photograph as the untouched base layer, then overlays 3-5 tiny chibi cartoon miniatures of the same person doing small playful actions, plus handwritten-marker captions, arrows, and sparkle doodles — like an illustrated selfie annotation.

When the agent picks it

Use when the user wants tiny cartoon mini versions of themselves drawn on top of their real photo — "add mini versions of me", "chibi me on my photo", "cute doodles on my selfie", "annotate my photo with chibis", "Q版", or any request for cartoon overlays on a real image. This is NOT a photo-to-cartoon conversion; the photo stays photographic and only the overlay elements are drawn. For a full cartoon avatar, use doodle-avatar. For a sticker sheet, use sticker-pack. For a grid of illustrated panels, use doodle-collage or full-body-collage. For mood-word captions, use mood-captions.

The full skill file

Every skill is one SKILL.md — this is the exact file the agent reads.

src/mastra/skills/chibi-mini-me/SKILL.md

Chibi Mini-Me

Overlay tiny hand-drawn chibi cartoon miniatures of the user on top of their real unretouched photo, with handwritten-style annotations and doodle accents.

When this is the right skill

Pick this skill when the user wants cute illustrated mini versions of themselves drawn over their actual photo — the photo stays real, only the overlays are cartoon. This is the viral “Q版” selfie-annotation trend.

Choose a different skill instead when the user asks for:

a full cartoon avatar replacing the photo → doodle-avatar
a 3×2 grid of illustrated panels → doodle-collage
full-body action poses in a grid → full-body-collage
a die-cut sticker sheet → sticker-pack
mood-word captioned panels → mood-captions
a greeting card → gift
no photo at all → surprise-me

What to draw

The base layer is the user’s photograph — it must remain photographic and unretouched. Do not cartoonify the photo, do not apply filters, do not alter the subject’s real face or body.

On top of that photo, draw:

3–5 tiny chibi miniatures of the same person (recognizable by hair, outfit colour, face shape) performing small playful actions that relate to the photo’s context — cheering, holding a mini sign, climbing on the subject’s shoulder, napping, drinking water, etc.
Short handwritten-marker captions and arrows pointing to elements in the photo or reacting to them. Keep text to 2–4 words per caption.
Small doodle accents — sparkles, hearts, stars, motion lines — scattered lightly so the overlay feels spontaneous, not cluttered.

The chibi minis use a bold-outline flat-colour naive doodle style (matching the Doodle AI house style) while the photo beneath stays photographic. This contrast is the defining characteristic of the skill.

How to run it

Confirm a photo is attached. If none is, ask the user to attach one — do not call the tool without a photo.
Call generateDoodle with skill: "mini-me" and the uploaded photo as imageUrl.
Report the result briefly. The app displays the image itself, so never paste the URL back into the conversation.