<div align="center">

Reconstruct a specific photographed subject and relationship as a modernist illustration that remains recognisable

<a href="README.md">简体中文</a> · <strong>English</strong> · <a href="README.ja.md">日本語</a> · <a href="README.ko.md">한국어</a> · <a href="README.ar.md">العربية</a>

</div>

Original Samples from X

<table> <tr> <td width="50%"><a href="https://x.com/xiaoxiaodong01/status/2090144142366233008"><img src="./assets/examples/sample-01.jpg" alt="XXD Panel 019 sample 1"></a></td> <td width="50%"><a href="https://x.com/xiaoxiaodong01/status/2090144142366233008"><img src="./assets/examples/sample-02.jpg" alt="XXD Panel 019 sample 2"></a></td> </tr> <tr> <td width="50%"><a href="https://x.com/xiaoxiaodong01/status/2090144142366233008"><img src="./assets/examples/sample-03.jpg" alt="XXD Panel 019 sample 3"></a></td> <td width="50%"><a href="https://x.com/xiaoxiaodong01/status/2090144142366233008"><img src="./assets/examples/sample-04.jpg" alt="XXD Panel 019 sample 4"></a></td> </tr> </table>

16:9 Left–Right Samples

<table> <tr> <td width="50%"><img src="./assets/examples/sample-05.png" alt="XXD Panel 019 Sample 5"></td> <td width="50%"><img src="./assets/examples/sample-06.png" alt="XXD Panel 019 Sample 6"></td> </tr> <tr> <td width="50%"><img src="./assets/examples/sample-07.png" alt="XXD Panel 019 Sample 7"></td> <td width="50%"><img src="./assets/examples/sample-08.png" alt="XXD Panel 019 Sample 8"></td> </tr> </table>

3:4 Top–Bottom Samples

These four works reuse the same four independent sources as the 16:9 set above, now regenerated as complete 3:4 top–bottom canvases. The original photograph remains above; the lower design follows Panel 019's original brief. English copy is generated only from each photograph's content, mood, or implication, and every result was individually reviewed.

<table> <tr> <td width="50%"><img src="./assets/examples/sample-09.png" alt="XXD Panel 019 additional top-bottom sample 1"></td> <td width="50%"><img src="./assets/examples/sample-10.png" alt="XXD Panel 019 additional top-bottom sample 2"></td> </tr> <tr> <td width="50%"><img src="./assets/examples/sample-11.png" alt="XXD Panel 019 additional top-bottom sample 3"></td> <td width="50%"><img src="./assets/examples/sample-12.png" alt="XXD Panel 019 additional top-bottom sample 4"></td> </tr> </table>

New sample: the four photos from the corresponding Chenxiang poster group were used in filename order and rendered with this Panel's original brief as a 16:9 left-right composition with original-brief-generated English copy.

Xiaoxiaodong (@xiaoxiaodong01) · 19 August 2026<br>

GPT2 × retro × flat × aesthetic prompt × VOL.019<br>

The post also demonstrates language preference: naming a target such as Spanish lets the copy transcreate naturally for that audience.

<p align="center"><a href="https://x.com/xiaoxiaodong01/status/2090144142366233008">View the original post and full prompt →</a></p>

These samples demonstrate the 019 aesthetic motive; they do not turn the post's earlier canvas into a current default. The four modes still follow the explicit pre-generation canvas and custom sizing logic below.

RECOGNISE FIRST · REDUCE WITH INTENT · COMPOSE WITH TYPE

XXD Panel 019 is an image-generation skill for Codex and compatible agents. It translates subject, posture, direction, proportion, and inseparable relationships into recognisable retro-modernist flat illustration. It is neither vector tracing nor an excuse to erase the source under generic geometry.

Reduction should remove noise, not identity.

<!-- xxd-human-intro:start -->

Where this Panel fits — and what it solves

When a photograph needs to become a poster, cover, social asset, or wallpaper, the real design problem is rarely a simple change of style. The challenge is to redesign it while preserving identity, key relationships, mood, and recognisability.

Panel 019 is suited to work that must retain the source's essential content while pursuing this visual direction: Reconstruct a specific photographed subject and relationship as a modernist illustration that remains recognisable. It begins with intended use and desired outcome; layout, size, text, and delivery format follow from those needs.

Use it when

• Content must remain anchored: people, objects, actions, and relationships should not be replaced by an unrelated template.
• The image needs stronger expression: you want a clearer editorial language, artistic treatment, or narrative focus.
• Delivery must stay flexible: one source needs comparison layouts, design-only artwork, multiple ratios, or a wallpaper set.

A simple way to begin

You do not need to review every option first. Send your Agent an image and say:

Use XXD Panel 019 on this image and recommend t

…