portfolio-detail — Chinese Atelier / Lishu Portfolio (with Secondary Detail Pages)
One-Liner Feature
Horizontal text rows + hover reveal on the right for large images + Alimama DaoLiTi artistic large characters + black and white pure color + light/dark + hash-routed detail pages + list/album dual view.
Differences from Other Portfolio Skills
| skill | core mode | who it's for |
|---|---|---|
portfolio-gen | infinite drag canvas + tile tiling | Photographers / graphic designers displaying samples |
portfolio-museum-gen | hero carousel + large cards + waterfall flow + detail pages (visual curation feel) | Video/installation/installation artists |
industrial-archive-gen | black framed flip book cover + sidebar archive + Nordic industrial | Physical products / artifacts / hardware |
lishu-portfolio-gen | Horizontal text rows + hover displays right thumbnail + lishu large characters + Chinese restraint | Bookbinding / typography / craftsmanship / Chinese atelier |
If the user's project falls into categories like "book / typography / exhibition / still life / Chinese herbal medicine / tea / pottery / wood / paper / incense" that require a restrained and elegant aesthetic — use this one.
When to Trigger
User says:
- "Create a Chinese / Eastern / classical style portfolio"
- "Create a lishu-style website / studio homepage"
- "Create a baiwu-style portfolio" / "Create a horizontal row list website like Gurgel D'Alfonso"
- "lishu-portfolio-gen ..."
And provides:
- studio description (required): name, lead, city, discipline, titles and stories for several works
- work images (optional): URL / upload / path. Can run without them (uses picsum placeholders)
If the user gives absolutely no work titles → you must ask at least which pieces are to be displayed. If the user only gave titles without details → generate them yourself: for each work, generate 1 lead sentence + 3 prose paragraphs + 6-8 spec lines + 3-6 picsum images + a few credits. The tone should match the type of work (book → binding / character → typography / object → still life / exhibition → exhibition visuals).
Two Operation Modes
Mode A · Claude Code / Local (with filesystem)
- Use
Bashto list folders,sipsto probe image dimensions - Output location:
~/Desktop/{slug}-portfolio.html(single file, all images via URL) - If the user gave local images: copy them to
~/Desktop/{slug}-portfolio-images/and change src to relative paths
Mode B · GLM / Online Chat (no filesystem)
- User unzips and sees 4 files
- Output only one complete HTML string, wrapped in a ````html` code block
- User uploads images but didn't give URLs → use
https://picsum.photos/id/{id}/{w}/{h}placeholders (use known existing IDs) - Do not attempt to access local files, do not suggest command-line operations
Rule of thumb: If you cannot Read and see local images → default to B.
File Division
| File | Role | How to Use |
|---|---|---|
reference.html | Complete finished reference — Baiwu 6 works | See: list item phrasing, prose tone, spec vocabulary, JSON field density |
template.html | Empty skeleton — same CSS/JS/routing system, 2 example items + 20 SWAP markers | Build on it, output final HTML |
README.md | User-facing usage instructions | Model doesn't need to read |
SKILL.md | Workflow for the model (this file) | Must read |
Core rule: All modifications are based on template.html; learn how to write from reference.html (tone, word count, tag usage), but do not copy specific content.
Work Steps
1. Read the Two HTML Files
- First read
reference.htmlto understand "what it looks like complete" (especially the field density and prose tone of theworks-dataJSON) - Then read
template.htmlto find the 20 SWAP locations
Do not change JS / Do not change CSS / Do not change the detail page routing system / Do not change the font import. SWAPs are all in the HTML text and JSON data.
2. Parse User Input to Extract Information
Extract:
- ATELIER_NAME_ZH — Chinese name of the studio (2-4 characters best, e.g. "白屋" "素园" "墨记" "青庐" "拙轩")
- PINYIN — Pinyin (5-6 English letters, all caps: BAIWU / SUYUAN / MOJI)
- CHAR — Single character for the logo (usually the first character of the studio name: 白 / 素 / 墨 / 青 / 拙)
- CITY — City
- DIRECTOR_ZH / DIRECTOR_EN — Lead's Chinese name + Pinyin
- DISCIPLINE — 3 discipline terms ("装帧 · 字体 · 静物")
- TAGLINE — One-line positioning statement (20-40 characters, kaishu aesthetic)
- YEAR_FOUNDED — Founding year (can be in Ganzhi: 乙未冬 / 甲辰春)
For each work, extract:
- slug — URL ID, lowercase English hyphenated (
kong-shan-ji/wu-ming-shu) - title_zh — Chinese title (2-5 characters best, the more concise the more stylish)
- title_en — English subtitle (book title style:
An Anthology of Empty Mountains) - year — Year or Ganzhi year (
甲辰 / 2024/2025/乙巳 / 2025) - tag — Top category, wrapped in square brackets, e.g.
[ 诗集 · 装帧设计 ]/[ 字体设计 · 中草药图谱 ] - cover — Detail page large image URL
- list row location —
City, Year · Craft/Spec(revealed on hover) - meta — 4 key-value pairs (detail page 4-column summary)
- lead — 30-70 character opening large text, can use
<em>keyword</em>to emphasize one point - prose — 3-4 prose paragraphs, each 80-150 characters
- specs — 6-10 spec table rows (format/paper/binding/typeface/print run/printing/craft/price etc.)
- gallery — 3-6 images, layout choice of three:
g-wide/g-half/g-third - credits — 4-7 pairs of role/name
If the user only gives one sentence/didn't provide a story → make it up yourself: based on the work's title and type, write a lead and prose that match the work's character. The tone should reference reference.html — not marketing language, but a feeling of "museum wall-text + studio notebook" (restrained, specific, slow, with traces of handwork).
3. List Works and Probe Dimensions (Mode A only)
ls -la "{works_path}"
sips -g pixelWidth -g pixelHeight "{works_path}/{file}" 2>/dev/null
Get for each image: (filename, w, h, aspect). Prioritize portrait images for list thumbnails (4:3 container, portrait images look better).
4. Replace the 20 SWAPs
| SWAP | Location | Content |
|---|---|---|
| 1 | <title> | {ATELIER_NAME} {PINYIN} — {DIRECTOR} Works / Studio |
| 2 | meta description | A one-line site description |
| 3 | Curtain single character | Studio name's first character (e.g. "白") |
| 4 | Logo SVG inner character | Same as SWAP-3 |
| 5 | Logo SVG inner pinyin | {PINYIN} in caps |
| 6 | Main navigation li | Usually 3 items: Works / About / Contact |
| 7 | Brand block Chinese name | {ATELIER_NAME} with characters separated by spaces (e.g. 白 屋) |
| 8 | Brand block English | {PINYIN} Studio · {CITY} |
| 9 | Tagline | Kaishu one-line positioning, 20-40 characters, can use <br> for line break |
| 10 | List items (N items) | Each <a class="item" href="#work/{slug}"> containing num/project-name/en-name/location/photo-thumb |
| 11 | Gallery items (N items) | Same N works in a more compact grid — slug must match the list |
| 12 | About section title | 关 于 {name}<span class="en">About</span> |
| 13 | About paragraphs | 1-3 kaishu paragraphs (founding background / lead / design philosophy), each 60-120 characters |
| 14 | About facts table | 4-6 li items (Founded/Location/Discipline/Director/Hours) |
| 15 | Contact section title | 来 信<span class="en">Get in Touch</span> |
| 16 | Contact paragraphs | 1-2 paragraphs inviting correspondence + response time commitment |
| 17 | Contact facts table | 4-6 items (E-mail/Phone/Address/Instagram/WeChat Official Account) |
| 18 | Rotating credit | Site by {WHO} (usually the studio name) |
| 19 | Footer three blocks | © / ICP / Social links |
| 20 | works-data JSON | Complete detail data for N works (key!) |
5. Write the works-data JSON (Most Crucial Step)
Each work must have one entry in the JSON array, and the slug must exactly match the #work/SLUG in the list/gallery card hrefs.
JSON fields:
{
"slug": "kong-shan-ji",
"title_zh": "空山集",
"title_en": "An Anthology of Empty Mountains",
"year": "甲辰 / 2024",
"tag": "[ 诗集 · 装帧设计 ]",
"cover": "https://picsum.photos/id/143/1800/1100",
"meta": [
{"label": "Type / 类型", "value": "诗集装帧"},
{"label": "Year / 年份", "value": "2024"},
{"label": "Edition / 印量", "value": "首印 320 本"},
{"label": "Format / 开本", "value": "130 × 195 毫米"}
],
"lead": "三十六首关于山的诗。我们想做一本——读者翻开后,先听见<em>风声</em>,再读见字的书。",
"prose": [
"Paragraph one: Origin of poems / Design context. 80-150 characters.",
"Paragraph two: Key decisions in production — typeface, paper, binding choices.",
"Paragraph three: Final presentation / Feel imparted to the reader."
],
"specs": [
["Format / 开本", "130 × 195 毫米"],
["Pages / 页数", "144 页"],
["Paper / 纸张", "本色棉纸 90gsm"],
["Binding / 装订", "线装四目"],
["Typeface / 字体", "老宋(标题)· 楷体(正文)"],
["Printing / 印刷", "单色凸印"],
["Edition / 印量", "首印 320 本"],
["Price / 定价", "¥ 388"]
],
"gallery": [
{"src": "...", "layout": "g-wide", "caption": "封面"},
{"src": "...", "layout": "g-half"},
{"src": "...", "layout": "g-half"},
{"src": "...", "layout": "g-third"},
{"src": "...", "layout": "g-third"},
{"src": "...", "layout": "g-third"}
],
"credits": [
["设计 / Design", "周墨白"],
["手工装订 / Binding", "陈漱石(杭州·订书坊)"]
]
}
JSON gotchas ⚠️:
- Quotations in Chinese prose must use full-width
「...」or"..."(U+201C / U+201D), cannot use ASCII"(will break JSON) - Do not write actual newlines in JSON strings; to break paragraphs, add another item in the prose array
<em>word</em>in lead directly write the HTML tag — it will be rendered as innerHTML- Gallery
layoutmust be one of three:g-wide(full width) /g-half(half width) /g-third(one-third width)
6. Image Source Selection
- picsum placeholders: Use
id/Xformat instead ofseed/X(more stable). Known good IDs: 0, 20, 40, 51, 64, 76, 88, 91, 96, 119, 121, 130, 143, 145, 158, 164, 165, 175, 177, 180, 187, 195, 200 - For the same work's cover/gallery, use adjacent IDs for consistent color tones
- Do not use the
?grayscaleparameter (users want colorful images)
7. Chinese Atelier Naming Inspiration (if the user doesn't provide)
If the user says "make one up" → choose 2 characters that feel ancient:
- Object class: 白屋 / 素园 / 墨记 / 青庐 / 拙轩 / 听雨 / 一勺 / 半亩
- Word/line class: 素白 / 子衿 / 砚池 / 半山 / 寻常 / 浮云
- Avoid functional words like "工作室 / 设计 / Studio"; let the name itself be poetry
8. Video Support
If a work's cover is .mp4 / .webm / .mov, the detail page will automatically use <video autoplay muted loop> instead of <img>. The model just needs to provide the video path in the cover URL.
9. Report
Mode A: One-line report.
Generated ~/Desktop/baiwu-portfolio.html — 6 works, hover over rows to see thumbnails, click to enter detail pages. Light/Dark toggle at bottom-left, list/gallery toggle at top-center.
Mode B: Directly provide the complete HTML code block + one line of explanation.
Save the above as
index.htmland double-click to open. Hover your mouse over any work name to see the thumbnail appear on the right, click to jump to the detail page. To change images, replace the picsum URL with your own image address.
Must Retain (Cannot Change)
- Font import:
@font-faceloading Alimama DaoLiTi's woff2 URL (do not change the font source) - CSS variable system: light/dark dual theme
- All JS: theme toggle, view mode toggle, hover-row interaction, rotating credit, detail page hash routing
applyRoute(),renderDetail() <script id="works-data">node must be retained<main id="detail-root">must be retained (detail page mounting point)body[data-view="home/detail"]andbody[data-mode="list/gallery"]state system- Five fixed-position elements: #logo / #main-menu / .theme-toggle / #project-view / #svp-credit
Quick Error Check
- Slug mismatch: list item
#work/kong-shan-ji↔ gallery item#work/kong-shan-ji↔ JSON"slug": "kong-shan-ji"— all three must match exactly - JSON quotes: Quotes in Chinese content must be full-width 「」 or
"...", not ASCII" - Too many gallery items: Maximum 6 images per work; exceeding will mess up the layout
- Prose too long: Each paragraph 80-150 characters; over 200 characters reduces readability
- List/gallery out of sync: Must contain the same set of works (one-to-one correspondence)
- picsum seed: Use
id/X, notseed/X, to avoid random IDs hitting deleted images - Works < 3: Acceptable; minimum 2 works to run
- Works > 8: Suggest splitting into two batches; too many list rows will make the page long
- Drop cap: The
.detail__prose p:first-of-type::first-letteruses DaoLiTi — the first paragraph must start with Chinese characters (not<em>or English) - Removing picsum colors: Images should remain colorful; do not add
filter: grayscale()in CSS