Visual Designer logo

Visual Designer

Community
XuanRanL
visual-designer

Optimize-phase pass that restructures a humanized draft into scannable visual components (tables, cited-stat grids, quotations, callouts, glossary, TL;DR) using native markdown the scoped CSS styles. Substance over ornament; enforces a floor and a ceiling. Runs after humanizer/geo/linker, before render-lint.

Overview

PublisherXuanRanL
Repositoryloamwright-SEO-Skill
Skill namevisual-designer
Stars
52
Forks
9
Bundled files
Instructions only
LicenseApache-2.0
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.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by XuanRanL on GitHub. Read the source before you install it.

Installation

Install the Visual Designer 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.
    https://github.com/XuanRanL/loamwright-SEO-Skill/tree/main/subskills/optimize/visual-designer
  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/XuanRanL/loamwright-SEO-Skill.git /tmp/loamwright-SEO-Skill
mkdir -p .claude/skills
cp -r /tmp/loamwright-SEO-Skill/subskills/optimize/visual-designer .claude/skills/visual-designer
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Visual Designer 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 Visual Designer 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 Visual Designer 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.

Visual Designer (optimize phase)

Make the article scannable and well-designed without changing what it says. This is a presentation pass. See agents/visual-designer.md for the role and hard constraints, and references/style/visual-design-components.md for the exact native-markdown pattern of every component. This SKILL is the playbook.

Step 0 — Read the situation

bash
# density result (which components are missing / over-used), if the check already ran
cat memory/workspace/{task}/visual-density.json 2>/dev/null

Read draft.md and angle.json (format_id). The format decides which components fit.

Step 1 — Plan by format (substance first)

FormatAdd (in priority order)
pillarcomparison table (Types/Choosing), ## By the Numbers, 1 quotation, ## Glossary (What-is-X), a checklist (Getting started)
comparison / reviewup to 3 tables (quick / pricing / feature), 1 verdict quotation
how-to / buyers-guidenumbered framework, ≤2 > **Warning:** callouts on failure steps, 1 spec table
listicle / roundupa spec table or ## By the Numbers per item cluster, ≤1 callout per top pick
weekly-digest## At a Glance table, ## By the Numbers, callouts for opinion blocks
opinion1 real quotation, minimal boxes

Only add what genuinely fits the content. Do NOT invent a comparison that is not in the text.

⛔ Stat-grid value contract (hard gate — scripts/lint/stat_grid_check.py, v3.39.0)

In a ## By the Numbers list the bold lead becomes a large DISPLAY FIGURE inside a narrow card. It must be a short number: ≤16 chars · longest word ≤10 chars · starts with a digit/$/~/± · does not end with :. All descriptive words go AFTER the closing **.

markdown
❌ - **30% more chlorophyll** in shaded tencha        →  ✅ - **30% more** chlorophyll in shaded tencha
❌ - **$12,000 to $25,000+ per month** retainer band  →  ✅ - **$12k-$25k/mo** enterprise retainer band
❌ - **Notched Izod verdict:** TPU absorbs 4x         →  ✅ - **4x** the notched Izod impact energy of PLA

A phrase in the bold overflows the card and chops MID-WORD (project-foxtrot shipped chlorophyll as chloroph / yll; a 591-post survey found 107 of 365 stat values breaking across 8 projects). An item with no number is not a stat — 25 of those 107 defects were writers using the grid as a generic bold-led checklist. Use a checklist or prose instead. Verify before finishing: python -m scripts.lint.stat_grid_check --workspace {task_id} --json

Step 2 — Convert prose to components (native markdown only)

  • Comparison table — a real GFM pipe table (≤6 cols). Keep each [claim:cN] on its cell.
  • By the Numbers — a NEW ## By the Numbers section + a - **N unit** label [claim:cN] list of 3 to 6 numbers ALREADY in the draft. Never invent a number.
  • Quotation — lift an existing, already-cited authority statement into > **Per {source},** ... [claim:cN].
  • TL;DR — a NEW ## TL;DR near the top + a 1 to 3 sentence answer-first paragraph drawn from the article's own conclusion.
  • Glossary — a NEW ## Glossary + - **Term** — definition list for first-mention terms.
  • Callout — > **Warning:/Tip:/Note:** ... for one thing per section a reader must not miss.
  • Pull-quote — > ❝ line ❞ at MOST once per article, editorial only.

Step 3 — Respect the floor and the ceiling

  • Floor: the article needs ≥1 Tier-A substance component (table / stat grid / quotation) and a weighted score ≥ the minimum. A wall of text fails.
  • Ceiling (evidence-based, do not exceed): ≤1 decorative pull-quote; ≤~1 callout per section; never let decorative boxes outnumber substance. Over-design (many colored boxes) is a measured failure (banner blindness) — restraint wins.

Step 4 — Idempotency + safety

  • If a component is already present, do NOT duplicate it. Only add missing ones.
  • Never edit heading text/anchors, [claim:*] / (Author, Year) citations, the References section, or the signature. Never write raw HTML. Never add an em-dash.

Step 5 — Verify + write evidence

bash
python -m scripts.lint.visual_density_check --workspace {task} --json

Confirm passed: true (or that you added everything the content supports). Then write memory/workspace/{task}/visual-design-report.json with _generated_by:"visual-designer-subagent", components_added, components[], density_score. The orchestrator gates on that file.

Notes

  • render-lint (L1-L9) runs right after you — any raw HTML or orphan bold you introduce is caught there, so keep to clean native markdown.
  • Typed callouts and pull-quotes get their color/pull-quote styling from the publisher hook _tag_callouts_and_pullquotes (it reads the blockquote's bold label). You only write the > **Label:** ... markdown; the class is added at publish.

Frequently asked questions

What does the Visual Designer AI skill do?

Optimize-phase pass that restructures a humanized draft into scannable visual components (tables, cited-stat grids, quotations, callouts, glossary, TL;DR) using native markdown the scoped CSS styles. Substance over ornament; enforces a floor and a ceiling. Runs after humanizer/geo/linker, before render-lint.

Why use Visual Designer on TypingMind?

Because you install it once and use it with any model. Visual Designer 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 Visual Designer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/XuanRanL/loamwright-SEO-Skill/tree/main/subskills/optimize/visual-designer. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Visual Designer?

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 Visual Designer?

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

Is the Visual Designer AI skill free?

Yes. It is published on GitHub by XuanRanL under the Apache-2.0 license. 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 👇