Notion Illustration logo

Notion Illustration

Organization
vm0-ai
notion-illustration

Notion-editorial-style hand-drawn spot illustration. Black brush-pen ink on white, tapered confident strokes, solid-black curly hair, solid-black pants/shoes, 3/4 face turned toward viewer with closed-eye smile and soft nose hint, open breathing body outlines, and 1-3 supporting scene props + ambient marks that frame the moment. Trigger when user says /notion-illustration, asks for a "Notion-style illustration", "Notion spot illustration", or a new piece in this hand-drawn brush-pen Notion editorial style.

Overview

Publishervm0-ai
Repositoryvm0-skills
Skill namenotion-illustration
Stars
76
Forks
18
Bundled files
9
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 9 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by vm0-ai on GitHub. Read the source before you install it.

Installation

Install the Notion Illustration AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/vm0-ai/vm0-skills.git /tmp/vm0-skills
mkdir -p .claude/skills
cp -r /tmp/vm0-skills/illustration-template/notion-illustration .claude/skills/notion-illustration
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Notion Illustration in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Notion Illustration on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Notion Illustration is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

/notion-illustration - locked Notion-style spot illustration

This skill produces single-character spot illustrations in the Notion editorial hand-drawn brush-pen style. Loose playful sketch energy, tapered confident ink strokes, breathing-open contours, solid-black hair and pants — plus a small supporting scene that anchors the moment in a place (not a figure floating in white). Each invocation swaps only the scene/activity — every other style axis stays locked.

Prompt interpretation

The user will give a short, simple prompt — usually just the activity (e.g. "reading", "walking the dog", "making coffee"). It is your job to brainstorm and expand it into a rich, specific scene. Do not ask the user to flesh it out. Do not stop to confirm. Just imagine the moment vividly and run.

For each invocation:

  1. Take the user's short prompt as the seed. Treat it as a starting hook, not the whole brief.
  2. Brainstorm the moment. Picture a specific time of day, a specific mood, a specific small detail that makes this scene feel this scene and not generic. What is the character actually doing in their hands? What's around them? What's the temperature of the moment — sleepy, focused, playful, cozy?
  3. Compose the scene with 1-3 supporting props + 2-4 ambient marks (see playbook for inspiration, but go beyond it — make creative choices specific to this prompt). Don't crowd the figure — supporting props frame, they don't compete.
  4. Keep the final image brief style-locked. The only creative variation should be the scene/activity and the supporting details.

Bias toward concrete, lived-in details: a half-drunk mug, a kicked-off shoe, a single tossed sock, a bookmark sticking out, a window cracked open. The scene should feel like the artist was there.

Locked style axes (NEVER vary)

Line quality

  • Black brush-pen / ink brush on slightly textured paper
  • Tapered strokes — thick in the middle, thin at the ends, like a real calligraphic brush
  • Confident, slightly imperfect — visible pressure variation, organic wobbles, ink density shifts
  • Hand-painted feel, NOT digital-clean, NOT vector, NOT marker-uniform-thickness
  • Each line breathes — slight personality

Flowing floaty gesture quality (潇洒 — critical)

Lines should have a sense of MOTION and FLOW — streamlined, airy, gestural. Each shape is SUGGESTED by a few confident sweeping strokes, not constructed as a careful contour with little gaps. Think calligraphy, not outlining.

The key insight: instead of drawing a complete outline and then adding small breaks, draw only a few gestural sweeps that imply the shape. WHOLE SECTIONS of the contour are simply absent — not "missing with a visible small gap" but not drawn at all. The viewer's eye fills in the missing 30-50% because the few strokes that ARE there carry enough information.

What this means concretely:

  • A couch is not an outline with breaks — it's maybe 3-4 confident flowing strokes (one sweep for the back curve, one for the seat, one for an armrest) and large stretches of white where there'd "normally" be a contour
  • A body is not a closed silhouette — it's a few flowing gestural strokes (one sweep down the back/side, one for an arm, one for a leg) leaving entire body sections unoutlined
  • Lines have a sense of MOTION — they sweep through the space like the brush was moving fast, with momentum, almost like brush calligraphy
  • Streamlined flowing curves — long elegant sweeps, not short careful segments
  • Floaty, light, airy — the figure and props feel like they could lift off the page
  • Strokes carry energy — pressure variation creates the feeling of movement frozen mid-gesture

Anti-patterns to avoid:

  • ❌ Drawing a complete outline and then adding tiny "breaks" in it
  • ❌ Short careful segments that connect at joints
  • ❌ Geometric correctness — perfect rectangles, perfect curves, symmetric shapes
  • ❌ Treating each contour as a closed shape with optional gaps
  • ❌ "Cute clean cartoon" tidiness
  • ❌ Lines that feel static or careful

The aesthetic to aim for: calligraphy-meets-life-drawing — confident sweeping brush gestures that capture the essence of a shape with minimum strokes, lots of breathing white space, energetic flow throughout.

Open breathing outlines (applies to BOTH figure AND scene props)

Many lines simply do not close. This is a defining feature of the style — not a subtle touch. Every illustration should have multiple obvious places where a contour stops in midair, leaves a visible white gap, or floats independently of the shape it "should" belong to.

On the figure:

  • Collar wide open, no closed neckline
  • Shirt bottom does NOT seal to the pants — clear visible white gap at the waist
  • Arms float as separate strokes near the body, not connected to the torso outline
  • One sleeve cuff or pant cuff may end with a small visible gap
  • A line in the middle of a sleeve or shirt panel may simply BREAK with white showing through, even when there's no "joint" reason for it
  • Hair silhouette outer edge may have one small open spot

On scene props (just as important — props should NOT all be tidy closed shapes):

  • Bed frame outline breaks at a corner, doesn't fully close
  • Couch back has a small open gap where the curve "should" continue
  • Nightstand legs don't fully connect to the top surface
  • Pot outline has a small break
  • Table edge has a tiny gap
  • Furniture corners are loose, not perfectly joined

The rule of thumb: at least 3-5 obvious visible breaks across the whole illustration (figure + props combined), in spots that look natural for a brush to have lifted. The viewer's eye mentally connects the strokes — the figure and scene still read clearly, but each contour was drawn as a confident gesture, not a careful closed-shape exercise.

Face (target: warm, inviting, facing viewer)

  • 3/4 angle turned toward the viewer (NOT strict side profile)
  • Both eyes visible as small upward-curved closed-smile arcs (⌒⌒) — closed-eye happy expression
  • Nose: tiny soft hint integrated into the face — almost no protrusion, NOT a triangular bump, NOT sharp
  • Mouth: small soft smile curve
  • Head slightly larger than realistic, soft round cheek

Hair

  • Solid black filled silhouette with mostly bumpy organic outer edges
  • Mostly solid — only ONE OR TWO very subtle internal brush separations (no busy white scribbles, no excessive gaps)
  • Voluminous shoulder-length curls, drawn loosely

Black silhouette fills

  • Pants drawn as a solid black silhouette with slightly brushy imperfect edges (not perfectly clean)
  • Shoes solid black
  • Selective use only — never the shirt, never the face

Figure proportions

  • Slightly cartoony — head a bit larger than realistic
  • Body loose and lightly elongated, NOT chibi, NOT abstract gesture, NOT anatomical
  • Limbs feel light and floaty, drawn as quick gestures

Scene context (NEW - always extend)

The figure should NEVER float alone in white space. Every illustration includes:

  • 1-3 supporting PROPS that frame the activity. Drawn in the same loose brush-pen style as the character — same line weight, same tapered quality, same level of detail. Props are quick supporting sketches, NOT detailed background. Examples:
    • Sleeping/waking → bed with rumpled blanket + pillow, nightstand with alarm clock, slippers on floor
    • Reading → couch with curved back + visible legs, side table, lamp, mug
    • Outdoor walk → small grass tufts under feet, a tree silhouette, a bird in the sky
    • Cooking → counter line, pot/pan, small ingredient marks
    • Working → desk with laptop, mug, papers, chair with curved back
    • Pet moment → pet rendered loose in same style, possibly on furniture
    • Celebration → cake/balloon/gift, plus floating sparkles
  • 2-4 ambient MARKS in the air around the scene that signal mood/motion. Tiny in scale, never crowding:
    • Sparkles / 4-point stars (✦ ✧ *) — fresh air, magic moment, celebration, morning
    • Action dashes (' ') near limbs — motion, stretch, swing, momentum
    • Hearts above heads — affection, comfort
    • Small sun/moon in a corner — time of day
    • Music notes — listening
    • Z's — sleep
  • Composition with the scene: character can be slightly off-center with props arranged on one side. Generous white margins still. The scene reads as a small captured moment, not a crowded illustration.

Composition rules

  • Single primary character (occasionally two if the activity requires)
  • Pure white background — NEVER add a sky/floor/wall fill
  • Generous white margin all around
  • NO color, NO gray fill (line/silhouette only), NO shading, NO hatching, NO border frame, NO text

Mood

  • Warm, candid, everyday — a confident artist's quick brush sketch capturing one playful moment in its setting

SCENE BRAINSTORM PLAYBOOK

This is inspiration, not a template. Always go beyond these starter combos — pick details specific to the moment you're imagining.

Activity seedPossible props (pick 1-3)Possible ambient marks (pick 2-4)
Waking up / stretchingbed + rumpled blanket + pillow, nightstand with alarm clock, slippers, kicked-off blanket on floor, half-open curtainsun in corner, action dashes near arms, sparkles, single floating feather from pillow
Readingcouch with legs, side table + mug, stack of books, reading lamp, blanket draped over lappages-flutter mark, soft Z if dozing, sparkle, single dust mote
Walking a dog / petleash, dog or cat (loose style), small grass tufts, fire hydrant, fallen leafsmall bird overhead, action dashes near feet, paw prints, heart
Working / focuseddesk + laptop + mug, scattered papers, chair back, sticky notes, plant on deskfloating workflow boxes, light bulb above head, sparkle, focused-eye-twinkle
Making coffee / kitchencounter line, mug, kettle/french press, coffee bag, single beansteam wisps, sparkle, small heart, drip-drop marks
Plant care / houseplantplant pot, watering can, single dropped leaf, second plant in backgroundsparkles around plant, drifting leaf, sun, water-drop marks
Yoga / movementyoga mat outline, water bottle, towelbreath marks, sparkles, small sun, calm-circle
Celebrationcake + candle, gift box, balloonsparkles overhead, hearts, stars, confetti dots
Music / listeningheadphones, phone, speaker, vinyl recordmusic notes, sparkles, vibration dashes
Cookingpot/pan, ingredient marks, cutting board, chef knifesteam, sparkle, heart
Cuddling petpet, couch or floor cushion, pet toyhearts, Z's, sparkle
Outdoor adventuresmall grass tufts, tree silhouette, backpack, water bottlebird overhead, sun, wind dashes

Brainstorming heuristic when stuck:

  • ONE prop that grounds the activity (the obvious one)
  • ONE prop that adds personality/specificity (the unexpected one — the kicked-off shoe, the half-eaten snack, the bookmark)
  • 2-3 small ambient marks for mood (sparkles, dashes, sun, music note, heart)
  • ONE atmospheric detail (time of day, weather, mood signal)

Style brief structure

Generate a single-scene spot illustration in the Notion editorial hand-drawn style shown in the reference images. Match the locked style EXACTLY.

SUBJECT — character: {{ACTIVITY}}

SCENE — environment around the character (critical, do NOT leave the figure floating in white):
{{SCENE — list 1-3 supporting props and 2-4 ambient marks. Example for "stretching after waking up":
- A loosely drawn BED behind/beside her: simple rectangular bed frame outline with a tall pillow at the head and a tossed-aside blanket draped over the side (rumpled, just-got-out-of-bed feeling)
- A small NIGHTSTAND with a SIMPLE ROUND ALARM CLOCK on top
- A pair of SLIPPERS on the floor beside her bare feet
- 2-3 tiny SPARKLE/STAR marks (small four-point asterisks ✦ ✧) in the air around her head
- 2-3 tiny ACTION DASHES near her arms suggesting the stretch motion
- A small SUN graphic in one upper corner (loose circle with a few short rays)}}

All scene props rendered in the SAME hand-drawn brush-pen style as the character — same line weight, same tapered strokes, same loose confident quality. Props should feel like quick supporting sketches, NOT detailed background — they frame the moment without crowding.

LINE QUALITY:
- Black brush-pen ink strokes with TAPERED quality (thick in the middle, thin at the ends) — calligraphic, confident, hand-painted.
- Visible pressure variation, organic wobbles, slight ink density shifts. NOT digital-clean, NOT uniform-thickness marker lines.
- Lines breathe with personality.

FLOWING FLOATY GESTURE QUALITY (潇洒 — critical):
- Each shape is SUGGESTED by a few confident SWEEPING gestural strokes, NOT constructed as a careful contour with little gaps.
- WHOLE SECTIONS of contour are simply ABSENT — not "missing with a small visible gap" but not drawn at all. The viewer's eye fills in 30-50% of the implied shape.
- A couch is 3-4 flowing sweeps (one for the back curve, one for the seat, one for an armrest) with large stretches of white where contour would "normally" be — NOT a complete outline with broken bits.
- A body is a few flowing gestural strokes (one sweep down the back/side, one for an arm, one for a leg) leaving entire body sections unoutlined — NOT a closed silhouette with little breaks.
- Lines have a sense of MOTION — they sweep through space like the brush was moving fast, with momentum, almost like calligraphy.
- Streamlined, flowing, light, FLOATY — the whole illustration feels airy.
- ANTI-PATTERNS to avoid: complete outlines with tiny "breaks", short careful segments, geometric correctness, treating each contour as a closed shape with optional gaps, static feel.
- The aesthetic: CALLIGRAPHY-MEETS-LIFE-DRAWING — confident sweeping brush gestures capturing the essence with minimum strokes and lots of breathing white space.

OPEN BREATHING OUTLINES (critical — applies to FIGURE AND SCENE PROPS):
- MANY lines simply do not close. This is the defining feature of the style — not subtle. At least 3-5 obvious visible breaks across the whole illustration.
- On the figure: collar wide open, shirt does NOT connect to the pants (visible white gap at waist), arms float as separate strokes not joined to the torso, a sleeve or pant cuff ends with a gap, sometimes a line in the middle of a panel simply breaks with white showing through.
- On the scene props: bed frame outline breaks at a corner, couch back has a small open gap, nightstand legs don't fully connect to the top, pot/cup/table outlines have small breaks, furniture corners are loose. Props should NOT all be tidy closed shapes.
- The viewer mentally connects the strokes — the figure and scene still read clearly, but each contour was drawn as a confident gesture, not a careful closed-shape exercise.

FACE:
- 3/4 angle TURNED TOWARD THE VIEWER (NOT strict side profile).
- BOTH EYES visible as small upward-curved closed-smile arcs — closed-eye happy expression, NOT dots, NOT open eyes.
- NOSE: a tiny soft hint integrated into the face, almost no protrusion. NOT triangular, NOT sharp.
- MOUTH: small soft smile curve.
- Head slightly larger than realistic, soft round cheek.

HAIR:
- Voluminous shoulder-length curly hair as a SOLID BLACK FILLED silhouette.
- Mostly solid — only ONE OR TWO very subtle internal brush separations. NO busy white scribbles, NO excessive gaps.
- Bumpy organic outer edge.

BLACK SILHOUETTE FILLS:
- Loose pants drawn as SOLID BLACK silhouette with slightly brushy imperfect edges.
- Shoes solid black (or barefoot small outlines).
- Shirt outline only with breathing breaks.

PROPORTIONS:
- Slightly cartoony — head a bit larger than realistic.
- Body loose and lightly elongated, limbs light and floaty.
- Quick gesture feel, NOT careful anatomical contour.

COMPOSITION:
- Character is the focal point, scene props arranged around her without crowding.
- Pure white background.
- Generous white margins, balanced composition (character can be slightly off-center).
- NO color, NO gray fill (line/silhouette only), NO shading, NO hatching, NO border frame, NO text.

MOOD: warm, candid, everyday — a quiet captured moment in its setting.

Match the locked exemplar style precisely. Do NOT copy any text or UI from references.

Style correction guide

If the output misses on a specific style axis, use a correction focused only on the visual issue:

IssueCorrection
Figure floating alone, no scene"Add scene context — 1-3 supporting props that frame the activity (specify which), plus 2-4 small ambient marks (sparkles, action dashes). All in the same brush-pen style."
Scene props too detailed / crowding"Simplify the scene — fewer props, looser brush sketches, more white space around the character."
Face turned to strict profile"Turn the face to 3/4 angle facing the viewer — show BOTH closed-smile eye curves, soften the nose to a tiny integrated hint."
Nose too triangular / sharp"Replace the pointed triangular nose with a soft gentle rounded curve barely interrupting the profile — almost integrated into the face line."
Lines too uniform / digital"Redraw with tapered brush-pen strokes (thick middle, thin ends) and visible hand-painted ink texture — pressure variation, slight organic wobbles."
Body lines all closed loops"Add visible white BREAKS in body outlines — collar wide open, waist completely unsealed, arm strokes float separately from torso."
Scene prop outlines too tidy / all closed"Break the scene prop contours too — bed frame doesn't close at a corner, couch back has an open gap, nightstand legs don't fully meet the top. Furniture and props should have at least 1-2 visible breaks each, same hand-drawn breathing quality as the figure."
Not enough overall breaks"More visible white breaks across the whole illustration — at least 3-5 obvious places where contours simply stop in midair. This is a defining feature of the style, not a subtle touch."
Hair too many white gaps"Reduce internal white gaps in hair — mostly solid black with only one or two subtle brush separations."
Feels too clean / designed"Looser playful 30-second-doodle quality, larger head proportions, brushy imperfect edges on the black silhouettes."
Lines feel stiff / static (not 潇洒, not floaty)"More FLOWING FLOATY gestural quality — each shape SUGGESTED by 3-4 confident sweeping brush strokes, NOT drawn as a complete outline with small breaks. WHOLE SECTIONS of contour simply absent (the back of the couch, half of one body side). Calligraphy-meets-life-drawing energy: minimum strokes, lots of breathing white space, motion and flow throughout. Viewer's eye fills in 30-50% of implied shape."

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Notion Illustration AI skill do?

Notion-editorial-style hand-drawn spot illustration. Black brush-pen ink on white, tapered confident strokes, solid-black curly hair, solid-black pants/shoes, 3/4 face turned toward viewer with closed-eye smile and soft nose hint, open breathing body outlines, and 1-3 supporting scene props + ambient marks that frame the moment. Trigger when user says /notion-illustration, asks for a "Notion-style illustration", "Notion spot illustration", or a new piece in this hand-drawn brush-pen Notion editorial style.

Why use Notion Illustration on TypingMind?

Because you install it once and use it with any model. Notion Illustration is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Notion Illustration in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vm0-ai/vm0-skills/tree/main/illustration-template/notion-illustration. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Notion Illustration?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Notion Illustration?

As many as you like. As long as a model supports skills, you can use Notion Illustration with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Notion Illustration AI skill free?

It is published on GitHub by vm0-ai. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇