Design-First UI Prompting Skill
This skill is for design-first prompting: turn fuzzy ideas into a tight spec that produces consistent UI.
When to use this skill
- Turning a fuzzy UI idea into a tight, skimmable spec using the GOAL / FORMAT / LAYOUT / TYPE SYSTEM / COLOR+MATERIAL / IMAGERY / COPY / CONSTRAINTS / NEGATIVE PROMPT skeleton.
- Locking a single "system" (layout, hierarchy, copy) first, then iterating by changing one variable at a time (angle/crop, accent color, card arrangement, background tone).
- Working around unreliable AI-generated typography with the 2-pass workflow: generate without text, then typeset the copy separately (e.g., in Figma).
- Building a "constraints card" (FONT / STYLE / MODE) to anchor a model's output like a mini style guide, or maintaining a local, gitignored reference pack instead of expecting the model to "remember" taste.
- Running the fast-iteration checklist (spacing, contrast, one-hero-line hierarchy, single accent color, texture) or the vague-user question list to clarify what a UI generation request actually needs.
- Not for defining or governing a shared design-token/visual-language system across pages and products (route to
design-system), not for responsive/breakpoint layout strategy or verification (route to responsive-design), and not for reusable component API architecture (no dedicated owner skill currently exists in this repo — treat as a design-system boundary question).
Core principle
Prompt like a design system, not a wish.
Prompt Structure (copy/paste)
Use this skeleton, then fill the blanks.
GOAL
- What are we making? (e.g., landing page hero / onboarding / dashboard / carousel slide)
- Who is it for? (persona)
- What’s the success criteria? (clarity, conversion, vibe)
FORMAT
- Size/aspect: (e.g., 1080x1350)
- Safe margins: (e.g., 90px)
LAYOUT (wireframe in words)
- Grid: (e.g., Swiss 6-col)
- Placement: (e.g., type-left / image-right)
- Hierarchy: H1 → subhead → body → CTA
TYPE SYSTEM
- Font vibe: (e.g., Söhne / Neue Haas / SF Pro)
- Weights: (H1 700, body 400)
- Leading: (tight for H1, readable for body)
- Tracking: (micro labels wider)
COLOR + MATERIAL
- Background: (hex or description)
- Text: (white/ivory/charcoal)
- One accent only: (cyan/lime/purple)
- Texture: (subtle grain, no plastic HDR)
IMAGERY / UI STYLE
- UI style: (minimal / glass / editorial / playful 3D)
- If photo: lighting + crop + texture rules
- If 3D: materials + lighting + softness
COPY (render EXACTLY)
- Line 1:
- Line 2:
- ...
CONSTRAINTS (change 1–2 things only)
- FONT: ___
- STYLE: ___
- MODE: ___
NEGATIVE PROMPT
- No logos, no watermarks
- No extra text beyond provided lines
- No gibberish typography
Rules that improve consistency
1) Lock one “system”, then iterate with variants
- First output: nail layout + hierarchy + copy.
- Variants: change ONE variable at a time:
- angle / crop
- accent color
- card arrangement
- background tone
2) Treat typography as fragile
If the model keeps misspelling:
- Use 2-pass workflow:
- Generate without text (reserve a clean text-safe area)
- Typeset in Figma
3) Use “constraints cards”
When you want the model to obey a style:
- Add a small “Constraints” panel with explicit values.
- It anchors the output like a mini style guide.
Example:
Constraints
FONT CANELA
STYLE MINIMAL
MODE DARK
4) Keep a local reference pack
Don’t ask the model to “remember” taste.
- Save references into a gitignored local reference folder, such as
refs/...
- Point prompts to the reference style
Fast iteration checklist (what to tweak)
- Spacing: margins, leading, baseline rhythm
- Contrast: background vs text
- Hierarchy: one hero line, one support line
- One accent only (don’t rainbow)
- Texture: add grain, remove smoothing
Questions to ask (when user is vague)
- What’s the single message of this screen?
- What’s the hierarchy (H1 / sub / CTA)?
- Which style lane: minimal editorial vs playful 3D vs glass UI?
- Any must-keep constraints (font vibe, color, spacing, grid)?
References
1---2name: design-first-ui-prompting3description: >4 Use when you need design-first, spec-driven, skimmable prompts for UI generation. Covers5 prompt structure, constraints, variations, typography/spacing rules, and iteration workflow6 for consistent UI outputs.7allowed-tools: Bash Read Write Edit Glob Grep WebFetch8compatibility: Prompt-authoring workflow; works with any UI-generating agent or model. No local dependencies.9license: MIT10metadata:11 tags: design, first, prompting, mengto-skills, ui-prompting, design-first12 version: "1.0"13 source: https://skillmd.com/skills/akillness/design-first-ui-prompting14 upstream: https://github.com/MengTo/Skills/tree/main/agent-skills/ui/design-first-ui-prompting15---1617# Design-First UI Prompting Skill1819This skill is for **design-first prompting**: turn fuzzy ideas into a tight spec that produces consistent UI.2021## When to use this skill2223- Turning a fuzzy UI idea into a tight, skimmable spec using the GOAL / FORMAT / LAYOUT / TYPE SYSTEM / COLOR+MATERIAL / IMAGERY / COPY / CONSTRAINTS / NEGATIVE PROMPT skeleton.24- Locking a single "system" (layout, hierarchy, copy) first, then iterating by changing **one variable at a time** (angle/crop, accent color, card arrangement, background tone).25- Working around unreliable AI-generated typography with the 2-pass workflow: generate without text, then typeset the copy separately (e.g., in Figma).26- Building a "constraints card" (FONT / STYLE / MODE) to anchor a model's output like a mini style guide, or maintaining a local, gitignored reference pack instead of expecting the model to "remember" taste.27- Running the fast-iteration checklist (spacing, contrast, one-hero-line hierarchy, single accent color, texture) or the vague-user question list to clarify what a UI generation request actually needs.28- Not for defining or governing a shared design-token/visual-language system across pages and products (route to `design-system`), not for responsive/breakpoint layout strategy or verification (route to `responsive-design`), and not for reusable component API architecture (no dedicated owner skill currently exists in this repo — treat as a `design-system` boundary question).2930## Core principle31**Prompt like a design system, not a wish.**3233## Prompt Structure (copy/paste)34Use this skeleton, then fill the blanks.3536```text37GOAL38- What are we making? (e.g., landing page hero / onboarding / dashboard / carousel slide)39- Who is it for? (persona)40- What’s the success criteria? (clarity, conversion, vibe)4142FORMAT43- Size/aspect: (e.g., 1080x1350)44- Safe margins: (e.g., 90px)4546LAYOUT (wireframe in words)47- Grid: (e.g., Swiss 6-col)48- Placement: (e.g., type-left / image-right)49- Hierarchy: H1 → subhead → body → CTA5051TYPE SYSTEM52- Font vibe: (e.g., Söhne / Neue Haas / SF Pro)53- Weights: (H1 700, body 400)54- Leading: (tight for H1, readable for body)55- Tracking: (micro labels wider)5657COLOR + MATERIAL58- Background: (hex or description)59- Text: (white/ivory/charcoal)60- One accent only: (cyan/lime/purple)61- Texture: (subtle grain, no plastic HDR)6263IMAGERY / UI STYLE64- UI style: (minimal / glass / editorial / playful 3D)65- If photo: lighting + crop + texture rules66- If 3D: materials + lighting + softness6768COPY (render EXACTLY)69- Line 1:70- Line 2:71- ...7273CONSTRAINTS (change 1–2 things only)74- FONT: ___75- STYLE: ___76- MODE: ___7778NEGATIVE PROMPT79- No logos, no watermarks80- No extra text beyond provided lines81- No gibberish typography82```8384## Rules that improve consistency8586### 1) Lock one “system”, then iterate with variants87- First output: nail **layout + hierarchy + copy**.88- Variants: change **ONE variable** at a time:89 - angle / crop90 - accent color91 - card arrangement92 - background tone9394### 2) Treat typography as fragile95If the model keeps misspelling:96- Use **2-pass workflow**:97 1) Generate without text (reserve a clean text-safe area)98 2) Typeset in Figma99100### 3) Use “constraints cards”101When you want the model to obey a style:102- Add a small “Constraints” panel with explicit values.103- It anchors the output like a mini style guide.104105Example:106```text107Constraints108FONT CANELA109STYLE MINIMAL110MODE DARK111```112113### 4) Keep a local reference pack114Don’t ask the model to “remember” taste.115- Save references into a gitignored local reference folder, such as `refs/...`116- Point prompts to the reference style117118## Fast iteration checklist (what to tweak)119- Spacing: margins, leading, baseline rhythm120- Contrast: background vs text121- Hierarchy: one hero line, one support line122- One accent only (don’t rainbow)123- Texture: add grain, remove smoothing124125## Questions to ask (when user is vague)126- What’s the single message of this screen?127- What’s the hierarchy (H1 / sub / CTA)?128- Which style lane: minimal editorial vs playful 3D vs glass UI?129- Any must-keep constraints (font vibe, color, spacing, grid)?130## References131132- Upstream source: [MengTo/Skills — design-first-ui-prompting](https://github.com/MengTo/Skills/tree/main/agent-skills/ui/design-first-ui-prompting)133- [ARTICLE.md](./ARTICLE.md) — longer-form article version of this workflow134- [REFERENCES.md](./REFERENCES.md) — external inspiration/reference sources (Mobbin, Bento Grids, Figma Community, Iconify, Simple Icons)