Iberian Vignette logo

Iberian Vignette

Organization
vm0-ai
iberian-vignette

Hand-drawn editorial vignette illustration in a locked Iberian-Mediterranean style — loose wobble ink linework, flat muted color fills with a cream ground and burgundy/wine as the anchor color, exactly one secondary accent (sage / mustard / dusty blue / terracotta), minimal abstract faces (tiny dot eyes, soft nose hint, almost no mouth), casually-observed body language, and a short Spanish serif lowercase wordmark anchored at the bottom of the canvas. Always grounded by an architectural / scenic prop (doorway, archway, balcony, courtyard, counter). Trigger when the user says /iberian-vignette, asks for an "Iberian vignette illustration", a "Spanish editorial illustration", a "Mediterranean observed-moment scene", or briefs with a scene metaphor + cast + complexity in this hand-drawn editorial style.

Overview

Publishervm0-ai
Repositoryvm0-skills
Skill nameiberian-vignette
Stars
76
Forks
18
Bundled files
4
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.

  • 4 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 Iberian Vignette 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/iberian-vignette .claude/skills/iberian-vignette
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Iberian Vignette 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 Iberian Vignette 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 Iberian Vignette 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.

/iberian-vignette — locked Iberian editorial vignette style

This skill produces quietly cinematic, observed-moment vignettes in a hand-drawn Iberian-Mediterranean editorial aesthetic. Every piece reads as the same illustrator's hand — same loose wobble ink, same muted cream-and-burgundy palette, same minimal abstract faces, same intimate observed posture, same short Spanish serif wordmark anchored at the bottom.

The style was trained from a Spanish "equipo" editorial cover and refined across both trade-shop scenes (cafe, bookshop, bakery, tailor, flower market, kitchen, bodega, ceramics, bicycle repair) and domestic / leisure moments (morning coffee on a balcony, a couple slow-dancing, three generations on a courtyard bench). See ref-l1-manana.png, ref-l2-ritmo.png, ref-l3-familia.png, and ref-l2-cafe.png for the visual target.

How to use this skill

The user gives a short brief — sometimes just a scene ("two friends at a market"), sometimes a fuller brief with palette + cast + wordmark. Your job:

  1. Map the brief to the six dials below. Fill in any missing dial with a tasteful, period-coherent default.
  2. Pick the matching complexity-tier reference anchor (L1 → ref-l1-manana.png, L2 → ref-l2-ritmo.png or ref-l2-cafe.png, L3 → ref-l3-familia.png) and feed it to the image model as a reference / source image.
  3. Compose a prompt from the Style brief structure below, preserving the locked axes verbatim and filling in the dial fields.
  4. Generate at portrait 2:3 (1024×1536 recommended) with the host style overlay bypassed so the model honors the locked brief.
  5. Inspect the output against the Quality checks list before returning it.

If the brief is empty, default to: solo woman in a soft cream linen outfit on a small Mediterranean balcony with a steaming mug, sage green herb pot on the railing, terracotta rooftops below, wordmark mañana.

Locked style axes (NEVER vary)

Medium and surface

  • Hand-drawn editorial ink + flat color — loose ink linework with visible wobble and varied weight (medium-fine pen feel; confident but imperfect, never digitally clean).
  • Flat opaque color fills — no gradients, no airbrush, no rendered shading inside shapes, no halftone, no paper texture.
  • Background: a single warm cream / off-white solid ground filling the entire canvas.
  • Floor (when shown): soft brown wood or stone.
  • Skin tones: warm sand / beige.

Anchor color

  • Burgundy / wine must be present in every piece — somewhere visible in the scene (a sweater, a dress, a hanging sign, a chair, a clay pitcher, an apron, a pot). Without it, the piece drifts away from the locked family.

Secondary accent (exactly one per piece)

  • Sage green · mustard yellow · dusty muted blue · terracotta-rust. Used sparingly on a single prop, plant, sign, tile, or one piece of clothing. Never two accents in the same piece — that breaks the calibrated mood.

Faces

  • Minimal abstract faces — tiny dot or short-mark eyes, soft nose hint (a single short curve), barely any mouth (a faint short line at most). Never detailed facial features. Never cartoon dot pupils or big anime eyes.
  • Pink blush is not used here (that belongs to other styles).

Hair

  • Flat solid color masses; the silhouette is one color block.
  • A few simple texture marks (a curl, a wave) are okay; never rendered strands or airbrushed highlights.

Posture and body language

  • Casually-observed — leaning, holding, gently propped, mid-action without performance. People stand, lean, sit, pour, wipe, kneel, dance, look out — never stiffly posed and never freeze-framed dramatically.
  • Energy goes in posture, never the face.

Composition

  • Portrait orientation, 2:3 (recommended 1024×1536).
  • Centered, cinematic, intimate — figures clustered, never sparse.
  • Always grounded by an architectural / scenic prop: a doorway, stone archway, window, balcony, courtyard, stone bench, counter, stoop, interior corner. Without the architectural anchor, the piece feels untethered.

Wordmark

  • A short lowercase serif noun anchored at the bottom center of the canvas. Bold serif. Always present.
  • Typically Spanish or another Iberian language: mañana, ritmo, familia, cafe, vino, pescado, sastre, libros, flores, cocina, bicis, panadería, ceramica, cena, tarde.

Mood

  • Calm, intimate, observed; quietly cinematic. Mediterranean afternoon light. Never frantic, never glamorous, never melancholic. This is the warmth of an everyday moment held still.

No edge artifacts

  • The original training reference was a magazine spread cropped to leave fragments of body copy (site, Sear) at the canvas edges. Do NOT include any cropped magazine text at the edges of the canvas — that was a training artifact, not part of the style.

Six creative dials (user-controlled)

#AxisWhat it controlsExample values
1Scene metaphorThe observed momentmorning coffee · evening dance · family courtyard · cafe conversation · flower market · tailor's workshop · bakery prep · bodega tasting · ceramics studio · bicycle repair · reading on a stoop · sharing a meal on a terrace
2CastWho is in the scenesolo (L1), pair (L2), 3–4 figures (L3). Mix genders. Vary ages — child, young adult, mature, elder all welcome
3Architectural propThe scenic anchorstone arched doorway · tall arched window · stone balcony · sunlit courtyard · wooden counter · interior corner with archway · stoop with cast-iron railing · open terrace with arbor
4Secondary accent (one only)The single non-burgundy color notesage green (plants, ceramics, signage) · mustard yellow (lamp, awning, sign, skirt) · dusty muted blue (tile work, pottery glaze, clothing) · terracotta-rust (bottles, amphora, pots, brick)
5WordmarkThe short Spanish noun anchored at the bottommañana · ritmo · familia · cafe · vino · pescado · flores · sastre · libros · cocina · cena · tarde · bicis
6ComplexityScene densityL1 — 1 figure + 1–2 props, single architectural framing element · L2 — 2 figures + small scene with arch prop · L3 — 3–4 figures + full vignette interior, courtyard, or storefront

Style brief structure (fill the bracketed fields, keep everything else verbatim)

Hand-drawn editorial vignette illustration in a loose-ink Iberian-Mediterranean style. Loose hand-drawn ink linework with visible wobble and varied weight. Flat muted color fills only — NO gradients, NO shading, NO texture. Warm cream off-white background filling the canvas, soft brown floor where applicable. Casually-observed body language — leaning, holding, gently propped, mid-action without performance.

[CAST: number + genders + ages + clothing notes; warm sand skin tones; one or more figures show BURGUNDY / WINE as the anchor color (sweater, dress, sign, prop, apron — pick one)]. Minimal abstract faces: tiny dot or short-mark eyes, soft nose hint, almost no mouth. Hair as flat color masses. [POSE: how the figures sit, lean, stand, hold, work, dance — must read as observed mid-moment, never stiffly posed].

The scene is grounded by [ARCHITECTURAL PROP: stone arched doorway / tall arched window / stone balcony / sunlit courtyard / wooden counter / archway / stoop / open terrace] framing the figures.

Secondary accent color: [SAGE GREEN / MUSTARD YELLOW / DUSTY MUTED BLUE / TERRACOTTA-RUST], used sparingly on [WHERE IT APPEARS — one prop or one piece of clothing]. NO second accent color anywhere.

Wordmark at the bottom center of the canvas: the word "[WORDMARK]" in bold serif lowercase.

DO NOT add any cropped magazine text at the edges of the canvas — keep the canvas clean. No additional text, no logos, no watermarks.

Portrait orientation 2:3, recommended 1024×1536.

Recommended model and reference handling

This style was trained on nano-banana-2 (ByteDance) in image-to-image mode, using one of the locked anchor references in this directory as the source image. The model is well-suited to the loose wobble ink + flat fill register and reliably preserves minimal-face anatomy across re-prompts when seeded from a reference.

When invoking the model:

  • Generate at portrait 2:3 (1024×1536 recommended). The wordmark needs the vertical room.
  • Bypass any host or app style overlay so the locked style brief above is what the model actually receives.
  • Pass the complexity-matched anchor as the source / reference image:
    • L1 (solo) → ref-l1-manana.png
    • L2 (pair) → ref-l2-ritmo.png (warm interior) or ref-l2-cafe.png (window seating)
    • L3 (3–4 figures) → ref-l3-familia.png
  • If the executor cannot fetch the reference attachment, document the limitation and proceed with a text-only prompt at higher creativity — outputs will be looser but still on-style.

Other models worth trying at lower fidelity: Seedream 5 (slightly cleaner edges, may over-tighten the wobble). Avoid Flux for this style — it tends to render the faces too photorealistically.

Quality checks (review before returning the image)

  • ✅ Line quality is loose and slightly wobbled, NOT crisp digital vector
  • ✅ Color fills are flat — no gradients, no shading inside shapes
  • ✅ Background is warm cream / off-white filling the canvas
  • ✅ Burgundy / wine appears somewhere on a figure or prop (anchor color present)
  • ✅ Exactly ONE secondary accent color (sage / mustard / dusty blue / terracotta) appears, used sparingly
  • ✅ Faces are minimal — tiny dot eyes, soft nose hint, almost no mouth; NOT detailed
  • ✅ Hair is a flat color mass, NOT rendered strands
  • ✅ Posture reads as observed and casual — leaning, holding, mid-action
  • ✅ The scene is grounded by an architectural / scenic prop (doorway, archway, balcony, courtyard, counter, etc.)
  • ✅ A short Spanish serif lowercase wordmark sits at the bottom center
  • ✅ NO cropped magazine text fragments at any canvas edge
  • ✅ Portrait 2:3 orientation; nothing crucial clipped at top or bottom

If any of these fail, re-prompt with stronger emphasis on the failed axis — move it earlier in the prompt and add an explicit anti-pattern instruction — rather than regenerating with the same prompt.

Anti-patterns to actively avoid

  • Crisp digital vector linework — wobble is the style; clean lines break it
  • Two or more accent colors — exactly one accent besides cream + burgundy + sand
  • Detailed facial features — minimal abstract faces only
  • Centered formal poses or symmetrical compositions — keep posture casual and observed
  • Empty / floating composition with no architectural anchor
  • Wordmark at the top, in a sans-serif, or in uppercase — bottom + lowercase + serif always
  • Cropped magazine text at the canvas edges (training artifact)
  • Adding aprons just because the original training image had aprons — aprons are scene-specific, not part of the locked frame

Example briefs

Example A — mañana (locked anchor, L1)

Brief: solo woman with morning coffee on a Mediterranean balcony, sage accent Filled:

  • Cast: 1 woman in a burgundy cardigan over cream linen, barefoot
  • Pose: standing at the balcony railing holding a steaming mug with both hands, gazing out at terracotta rooftops
  • Architectural prop: small stone balcony with curtain-draped balcony door behind her
  • Accent: sage green (potted herb on the railing)
  • Wordmark: mañana
  • Anchor: ref-l1-manana.png

Example B — ritmo (locked anchor, L2)

Brief: couple slow-dancing in a warm-lit living room, mustard accent Filled:

  • Cast: 1 man in burgundy crew-neck sweater + 1 woman in cream blouse and mustard skirt
  • Pose: slow-dancing close, his hand at her waist, her arm draped over his shoulder
  • Architectural prop: interior living-room corner with a vintage record player on a side table
  • Accent: mustard yellow (her skirt + the warm table-lamp glow)
  • Wordmark: ritmo
  • Anchor: ref-l2-ritmo.png

Example C — cafe (locked anchor, L2 alternate)

Brief: two friends sharing late-afternoon coffee near a tall arched window Filled:

  • Cast: 1 man in burgundy crew-neck sweater + 1 woman in cream blouse with mustard skirt
  • Pose: leaning gently toward each other over small espresso cups at a small wooden cafe table
  • Architectural prop: tall arched window beside the table
  • Accent: mustard yellow (small mustard table lamp on the cafe table)
  • Wordmark: cafe
  • Anchor: ref-l2-cafe.png

Example D — familia (locked anchor, L3)

Brief: three generations in a sunlit Mediterranean courtyard, dusty blue accent Filled:

  • Cast: grandmother with silver hair in a burgundy linen dress + mother in cream blouse and dark trousers + young child in a cream sundress holding an orange
  • Pose: all three on a stone courtyard bench, grandmother turned toward the child, mother leaning gently back
  • Architectural prop: arched stone courtyard wall with a small dusty-blue-tiled fountain and a potted lemon tree
  • Accent: dusty muted blue (fountain tile)
  • Wordmark: familia
  • Anchor: ref-l3-familia.png

Example E — vino (new brief, L3, terracotta accent)

Brief: three vintners around a tasting bar in a small wine bodega Filled:

  • Cast: older bearded man pouring + younger woman across the bar swirling a glass + younger man wiping a glass behind the bar
  • Pose: mid-pour, mid-swirl, mid-wipe — observed working moment
  • Architectural prop: wooden bar inside a small bodega with wooden shelves behind
  • Accent: terracotta-rust (rows of terracotta-colored bottles + a clay amphora at the end of the bar)
  • Wordmark: vino
  • Anchor: ref-l3-familia.png

Example F — flores (new brief, L2, mustard accent)

Brief: two flower-market vendors under a striped awning Filled:

  • Cast: 1 woman in burgundy apron arranging bouquets + 1 man in burgundy apron leaning on the side of the stall holding a stem
  • Pose: working the morning stall, mid-conversation
  • Architectural prop: wooden flower stall under a mustard-and-cream striped awning
  • Accent: mustard yellow (awning stripes)
  • Wordmark: flores
  • Anchor: ref-l2-ritmo.png

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 Iberian Vignette AI skill do?

Hand-drawn editorial vignette illustration in a locked Iberian-Mediterranean style — loose wobble ink linework, flat muted color fills with a cream ground and burgundy/wine as the anchor color, exactly one secondary accent (sage / mustard / dusty blue / terracotta), minimal abstract faces (tiny dot eyes, soft nose hint, almost no mouth), casually-observed body language, and a short Spanish serif lowercase wordmark anchored at the bottom of the canvas. Always grounded by an architectural / scenic prop (doorway, archway, balcony, courtyard, counter). Trigger when the user says /iberian-vignet...

Why use Iberian Vignette on TypingMind?

Because you install it once and use it with any model. Iberian Vignette 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 Iberian Vignette in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/vm0-ai/vm0-skills/tree/main/illustration-template/iberian-vignette. 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 Iberian Vignette?

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 Iberian Vignette?

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

Is the Iberian Vignette 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 👇