Cozy Parlor logo

Cozy Parlor

Organization
vm0-ai
cozy-parlor

Hand-painted watercolor + ink-line illustration style — one anthropomorphic animal in a quiet domestic interior, on clean cool-white watercolor paper (never amber-tinted), with a neutral cool palette and a single hot accent pop. Cozy, hushed, picture-book mood. Trigger when the user says /cozy-parlor, asks for a 'cozy parlor illustration', a 'watercolor animal scene', a 'picture-book interior', or a new piece in this neutral-palette watercolor style.

Overview

Publishervm0-ai
Repositoryvm0-skills
Skill namecozy-parlor
Stars
76
Forks
18
Bundled files
5
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.

  • 5 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 Cozy Parlor 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/cozy-parlor .claude/skills/cozy-parlor
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Cozy Parlor 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 Cozy Parlor 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 Cozy Parlor 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.

Cozy Parlor Illustration

Use this register-only image style when the user asks for a cozy watercolor animal scene, a domestic-interior picture-book illustration, a "/cozy-parlor" piece, or any new piece in this hand-painted-on-cool-white-paper neutral-watercolor style.

This style is distinct from folk-storybook (warmer aged-paper folk gouache, fully painted) and painterly-botanical (single-figure portrait with foliage). Cozy Parlor is a scene-based watercolor: clean cool paper, brushy black ink line over translucent washes, one anthropomorphic animal mid-quiet-activity, and a single hot accent pop against a cool palette.

Bundled reference imagery — four canonical outputs that demonstrate the framework in action:

  • ref-frog-letters.jpg — Frog at a writing desk dipping a quill, sage curtain + pale-blue stripe wallpaper + slate gingham blotter, mustard lamp accent. Demonstrates the Sage-Study palette and the L3 full-vignette archetype.
  • ref-mouse-baker.jpg — Mouse kneading bread on a marble counter, delft tile backsplash + polka-dot dishcloth + striped flour jar, cherry-red tulip accent. Demonstrates the Cornflower-Kitchen palette and the L2 corner-vignette archetype.
  • ref-hedgehog-records.jpg — Hedgehog in headphones curled on a tufted pouf, lavender wavy wallpaper + checkered floor + striped throw, magenta hibiscus accent. Demonstrates the Lavender-Lounge palette and L3 full-vignette at a moodier emotional register.
  • ref-otter-painter.jpg — Otter in a chalk-stripe smock painting at an easel, floral curtain + gingham smock pocket + checkered rug, cherry-red palette accent. Demonstrates the Mint-Studio palette and that the activity itself is the metaphor — never a generic editorial scene.

Locked Style Fundamentals (never vary)

AxisSpec
MediumHand-painted translucent watercolor washes + brushy black ink line drawn ON TOP
PaperClean cool-white watercolor paper — bright neutral white, NO amber tint, NO warm cream overlay, NO sepia wash
TextureVisible brushstrokes, watercolor pigment bleeds, faint paper grain
SubjectONE anthropomorphic animal character, mid-quiet-activity in a domestic interior
Character faceClosed crescent eyes + tiny triangle nose. Sometimes a soft pink cheek dot. No open eyes, no eyebrows, no visible mouth. Expression lives in posture, never in the face.
SettingCozy domestic interior — study, kitchen, parlor, greenhouse, studio, library, etc. Never outdoors, never abstract, never landscape.
Patterned surfacesAt least 2 surfaces in frame carry decorative pattern (wallpaper, rug, curtain, upholstery, tile, textile, ceramic, smock, cushion).
Palette leadCool/neutral — sage, slate-blue, cornflower, lavender, dove-gray, mint, chalk-white. Never lead with cream, butter, peach, amber, terracotta.
AccentExactly ONE hot pop (cherry-red, mustard, magenta-pink, terracotta) on a SINGLE OBJECT — a lamp, book, tulip, towel, etc. Never a wash, never a wall color.
StagingObject-rich personality props — books, plants, dishes, lamps, jars, tools of the activity. Lived-in feel.
CanvasPortrait 1024x1536 (2:3) via gpt-image-1.5, --quality high

The chill-face rule (critical)

Every character keeps closed crescent eyes + a soft content smile, even mid-action. A bear pouring boiling water keeps closed eyes. A mouse mid-knead keeps closed eyes. An otter mid-brushstroke keeps closed eyes. All energy lives in the body posture, the activity props, and the scene — never in the face. (Per [[feedback_illustration_chill_expressions]].)

The cool-paper rule (critical)

The paper is the cool anchor of the whole style. If the background reads cream, butter, beige, or amber — the piece is wrong. The watercolor must read neutral. Hot accents are welcome as a single object (the lamp, the tulip, the book). Hot accents as a wash or wall color break the style. (Per [[feedback_watercolor_neutral_paper]].)

The activity-is-metaphor rule

The scene metaphor is the activity itself, not a relabeled editorial template. A "writing" scene shows a real desk, real letters, a real quill — not a filing cabinet labeled "letters". A "baking" scene shows a real counter, real dough, a real oven. (Per [[feedback_illustration_scene_not_template]].)

The cast-rotation rule

Across a series, vary the animal. Don't use the same cat in every piece. Cast is a per-scene dial, not a locked mascot. Mix: cat, frog, mouse, hedgehog, otter, rabbit, fox, bear, capybara, owl, badger, raccoon… (Per [[feedback_illustration_character_variety]].)

Dials (vary per piece)

  • Cast — pick one animal from a wide pool; rotate across the series
  • Activity / scene metaphor — writing letters, baking, kneading dough, listening to records, reading, painting, watering plants, brewing tea, knitting, mixing potions, repairing a watch, drying herbs
  • Palette family — pick one of the canonical five, or compose a new cool/neutral lead:
    • Sage-Study: sage-green + dusty slate-blue + chalk-white → MUSTARD accent
    • Cornflower-Kitchen: dusty cornflower-blue + mint + chalk-white → CHERRY-RED accent
    • Lavender-Lounge: pale-lavender + dusty teal + chalk-white → MAGENTA-PINK accent
    • Mint-Studio: soft mint-green + dove-gray + chalk-white → CHERRY-RED accent
    • Slate-Parlor: dove-gray + slate-blue + chalk-white → CHERRY-RED accent
  • Pattern motif stack — pick 2–3 from: stripes, gingham, polka-dot, small florals, delft tile, wavy/scallop, checkered, lattice
  • Hot accent object — ONE object carries the hot color
  • Complexity
    • L1: close character + 1 prop, tight crop, 2 patterned surfaces max
    • L2: corner vignette, 2–3 patterned surfaces, 4–6 props
    • L3: full-room vignette, 3+ patterned planes, 6+ props

Prompt template

Hand-painted watercolor + brushy black ink line on CLEAN COOL-WHITE watercolor paper — no amber tint, no warm cream overlay, the paper itself reads bright neutral white. Cozy domestic interior. ONE anthropomorphic animal character with simple closed crescent eyes and a tiny triangle nose — minimal facial features, expression through posture. At least two patterned surfaces in frame. Object-rich personality staging. Visible brush strokes and translucent watercolor washes. Picture-book illustration mood. Portrait composition.

PALETTE (cool/neutral): {PALETTE_LEAD}, one HOT {ACCENT_COLOR} accent on {ACCENT_OBJECT}. Avoid cream, butter, peach, amber, terracotta. Overall wash cool and balanced.

SCENE: {CAST_DESCRIPTION} {ACTIVITY_DESCRIPTION}, in a {SETTING}. PATTERN MOTIFS: {PATTERN_1}, {PATTERN_2}, {PATTERN_3}. PROPS: {PROP_LIST}. COMPLEXITY: {L1 / L2 / L3 description}.

Model guidance

This style is designed for narrative scene compositions with multiple props, patterned surfaces, and a single anthropomorphic character — the kind of scene a strong text-to-image model can compose in one pass from a detailed prompt. Recommended pairing: gpt-image-1.5 at high quality, portrait 1024x1536, with prompt enhancement disabled so the locked-fundamentals language reaches the model verbatim.

The bundled reference images (ref-frog-letters.jpg, ref-mouse-baker.jpg, ref-hedgehog-records.jpg, ref-otter-painter.jpg) are canonical outputs of this framework — provided for human style study and prompt authoring. They are not required as image inputs at generation time — the style transfers entirely through the detailed prompt language.

If output drifts (paper reads cream/amber, palette unifies under a warm overlay, ink lines feel too vector-clean, character face opens up), tighten the prompt by re-emphasizing "CLEAN COOL-WHITE watercolor paper, no amber tint, no warm cream overlay, paper reads bright neutral white" and the closed-crescent-eye rule — rather than reaching for image-to-image.

Example briefs

Example 1 — Bear knitting (Sage-Study, L2)

PALETTE: sage-green + dusty slate-blue + chalk-white lead, one HOT MUSTARD accent on a ball of yarn.

SCENE: A small brown BEAR in a chunky cardigan sits in a wingback chair knitting a scarf, in a quiet LIVING ROOM CORNER. PATTERN MOTIFS: slate-blue chevron wallpaper, a small floral cushion, a checkered throw. PROPS: a basket of yarn balls, a half-finished scarf trailing, a wooden side table with a mug, a small dachshund-shaped doorstop. COMPLEXITY: L2 corner vignette.

Example 2 — Cat watering plants (Slate-Parlor, L1)

PALETTE: dove-gray + slate-blue + chalk-white lead, one HOT CHERRY-RED accent on a watering can.

SCENE: A tuxedo CAT tips a small red watering can over a potted spider plant on a windowsill, in a sun-cool NOOK. PATTERN MOTIFS: a slate-blue chalk-stripe curtain, a small polka-dot pot wrap. PROPS: a wooden stool, a single trailing pothos, a teacup on the sill. COMPLEXITY: L1 close-character composition.

Example 3 — Owl reading (Lavender-Lounge, L3)

PALETTE: pale-lavender + dusty teal + chalk-white lead, one HOT MAGENTA-PINK accent on a single peony in a vase.

SCENE: A small barn OWL nestled in an oversized velvet armchair reads a thick book by lamplight, in a quiet STUDY. PATTERN MOTIFS: lavender wavy wallpaper, a tufted chair pattern, a striped rug, a checkered footstool. PROPS: a leaning bookshelf, a tea tray with a flowered cup, a brass desk lamp, a single magenta peony in a vase, a stack of books on the floor. COMPLEXITY: L3 full-room vignette.

Anti-patterns (the deal-breakers)

  • ❌ Warm cream / amber / butter / peach paper background — paper must stay cool-neutral. (This is the #1 failure mode.)
  • ❌ Reusing the same animal cast across the series — every piece picks a fresh animal.
  • ❌ Generic editorial scene with relabeled props (filing cabinets, megaphones, abstract office shapes).
  • ❌ Open eyes, eyebrows, or visible mouth on the character — closed crescents only.
  • ❌ Outdoor scenes, landscapes, abstract backgrounds — always a domestic interior.
  • ❌ Only one patterned surface — must have at least two.
  • ❌ Hot accent used as a wall color, lighting tint, or background wash — always a single object.
  • ❌ Vector-clean line work or flat color fills — must feel hand-painted with visible brushstrokes.
  • ❌ Photorealism, 3D rendering, airbrush gradients — watercolor + ink only.

Series example — same style across a varied set

A four-piece set for blog covers in this style might include:

  1. Frog writing letters (Sage-Study, L3) — a study/quiet-focus scene
  2. Mouse baker (Cornflower-Kitchen, L2) — a kitchen/craft scene
  3. Hedgehog with headphones (Lavender-Lounge, L3) — a leisure/listening scene
  4. Otter painter (Mint-Studio, L2) — a creative/making scene

Each piece flexes cast + activity + palette + complexity against the same locked frame. The four feel cousin-related, never identical.

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 Cozy Parlor AI skill do?

Hand-painted watercolor + ink-line illustration style — one anthropomorphic animal in a quiet domestic interior, on clean cool-white watercolor paper (never amber-tinted), with a neutral cool palette and a single hot accent pop. Cozy, hushed, picture-book mood. Trigger when the user says /cozy-parlor, asks for a 'cozy parlor illustration', a 'watercolor animal scene', a 'picture-book interior', or a new piece in this neutral-palette watercolor style.

Why use Cozy Parlor on TypingMind?

Because you install it once and use it with any model. Cozy Parlor 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 Cozy Parlor in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vm0-ai/vm0-skills/tree/main/illustration-template/cozy-parlor. 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 Cozy Parlor?

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 Cozy Parlor?

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

Is the Cozy Parlor 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 👇