Kimi-K3
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.
A design skill that transforms real ideas into finished visuals by first understanding content, then selecting specific symbols, actions, and traces, and finally assembling image and typography into a coherent artifact.
11
No data
0
0
No data
2026-07-29
It shifts the focus from style to meaning, making AI-generated designs content-specific and human-like, with deliberate reading order and editable copy.
A design skill for turning a real idea into a finished visual with a point of view.
它不是给普通海报加一点纸张纹理,而是先理解内容,再找到具体的符号、动作和痕迹,最后把图像与准确文字组装成完整作品。
The output can be quiet, playful, cinematic or strange. It should still be understandable, content-specific and recognizably made by a person.
Most visual prompts begin with style. This skill begins with meaning.
It asks what enters, what changes, what comes out and what proves the change. It then builds a visual sentence from a recognizable subject, a specific action and a visible consequence. Handmade line, print texture, color and type come afterwards.
The finished artifact must include exact typography and a deliberate reading order. A generated illustration by itself is not treated as a finished poster.
With the Skills CLI:
npx skills add pepperliu722-sudo/create-quiet-humanist-designs --skill create-quiet-humanist-designs
Or copy skills/create-quiet-humanist-designs into your agent's skills directory.
Use $create-quiet-humanist-designs to turn this product feature into a 4:5
social poster. First explain the feature in plain language, then create three
visually different directions. Keep all final copy exact and editable.
调用 $create-quiet-humanist-designs,把这篇文章做成一张有具体符号和人物动作的
编辑海报。不要先套风格,也不要只做一张有氛围的插画。最后把标题、说明和署名完整排进去。
Use $create-quiet-humanist-designs to translate this campaign into an email
header, a presentation title slide and a web hero. Keep the idea coherent,
but redesign the composition for each medium instead of cropping one poster.
content model
↓
visual anchors
↓
anchor + action + trace
↓
three genuinely different directions
↓
image and exact typography assembled together
↓
full-size and thumbnail QA
The detailed rules live in SKILL.md. Supporting references are loaded only when the task needs them, so the core skill stays focused.
skills/create-quiet-humanist-designs/
├── SKI
Similar projects matched by category, topics, and programming language.
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.

A structured investing guide for Chinese beginners covering US stocks, options, and cryptocurrency, with focus on foundational concepts and risk awareness.

A curated collection of Computer Science fundamentals (PDFs, notes, cheatsheets, interview question banks) for placement preparation, covering seven core subjects plus general resources.