⏳ This skill is pending AI review.
Scores will appear once the review pipeline completes.
make-photo-stamp-archive
Transform one or more supplied photos into clean direct-splice archival artworks that pair a faithfully preserved photograph with a warm-white paper panel containing a compact hand-pressed custom seal or stamp. Use for requests mentioning 图章、印章、stamp、seal、照片加图章、照片档案、左右拼接、清爽纸张、圆形章、方形章、异形章、建筑轮廓章、主体缩小、移动到角落, or iterative changes to stamp shape, border, scale, position, ink color, caption, paper age, or splice orientation.
Choose how to use this skill
You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.
1. Native installer
This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.
Do not guess an installer command or replace an existing version without reviewing the diff.
2. Complete package recommended
Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.
No complete ProSkills package is published for this listing yet.3. Prompt-only
Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.
Need only the instruction file?
Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.
No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.
// RATINGS
Not yet listed on ClawHub or SkillsMP
// README
Make Photo Stamp Archive
Make Photo Stamp Archive 是一个 Codex Skill:它可以把一张或多张照片转化为安静、克制的档案拼接作品——一侧忠实保留原照片,另一侧使用暖白纸张与根据主体定制的手工图章,两块画面以笔直边界直接拼接。
调用名称为 make-photo-stamp-archive。
核心能力
- 忠实保留原照片中的人物、面孔、物件、建筑、文字、色彩关系与场景逻辑。
- 默认采用横向左右直拼,也可按要求改成上下拼接;主分界始终清晰、笔直。
- 根据主体设计圆形、方框、横向全景、拱形或自定义轮廓图章。
- 呈现干墨、磨损网点、压力不均、边缘缺墨与轻微套色偏差等真实手压质感。
- 支持只修改图章形状、边框、大小、位置、墨色、标题、纸张年代感或拼接方向。
- 多张照片会分别生成独立成品,不合并成拼贴画。
视觉系统
- **照片区:**忠实、真实,默认约占画布的 55%。
- **纸张区:**干净的暖白档案纸,带轻微纤维与扫描痕迹,默认约占画布的 45%。
- **图章组:**体量小、靠角落放置,约占纸张区的 30%。
- **留白:**纸张区约 70% 保持安静、空白。
- **文字:**在图章附近放置小号褪色打字机文字,不覆盖图章。
- **气质:**克制、平面、可触摸、档案化,像一段被保存的记忆。
最终结果始终是一张平面成品,不会做成书本样机、剪贴簿、前后对比板、缩小照片贴片或装饰性海报展示。
作例
| 作例 01 | 作例 02 |
|---|---|
| 作例 03 |
|---|
运行要求
- Codex,或其他兼容 Skill 的运行环境。
- 能够读取图片,以检查源照片。
- 能够生成或编辑图片,以制作和修改最终成品。
本 Skill 包不包含 API Key、外部字体、脚本或需要额外下载的运行素材。最终画质与源照片保真度取决于宿主环境可用的图像模型。
安装
git clone https://github.com/Dlcccc71913/skill-make-photo-stamp-archive.git
mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills"
cp -R skill-make-photo-stamp-archive \
"${CODEX_HOME:-$HOME/.codex}/skills/make-photo-stamp-archive"
如果 Skill 没有立即出现,请重启 Codex。
使用
使用 $make-photo-stamp-archive,把这张照片制作成带定制图章的档案直拼作品。
进行局部修改时:
保持照片区完全不变,只把图章缩小 10%,并移动到右上角。
仓库结构
.
├── SKILL.md
├── agents/
│ └── openai.yaml
├── references/
│ └── prompt-template.md
├── README.md
├── README.en.md
├── README.ja.md
└── LICENSE
许可证
// HOW IT'S BUILT
// FILES
7 files shipped with this skill. Inline text is the catalog copy. The upstream link is the source repository.
SKILL.md5.4 KBc243c69c4df3upstream ↗
--- name: make-photo-stamp-archive description: Transform one or more supplied photos into clean direct-splice archival artworks that pair a faithfully preserved photograph with a warm-white paper panel containing a compact hand-pressed custom seal or stamp. Use for requests mentioning 图章、印章、stamp、seal、照片加图章、照片档案、左右拼接、清爽纸张、圆形章、方形章、异形章、建筑轮廓章、主体缩小、移动到角落, or iterative changes to stamp shape, border, scale, position, ink color, caption, paper age, or splice orientation. --- # Make Photo Stamp Archive Create one finished raster composite per source photo: a faithful photographic panel directly joined to a quiet paper panel with a compact subject-specific seal. Read [references/prompt-template.md](references/prompt-template.md) before generating or revising an image. ## Workflow 1. Inspect every local source image with `view_image`. 2. Identify the subject, distinctive silhouette, original color memory, viewpoint, scene geometry, important text, objects, people, animals, occlusions, and visual weight. 3. Lock invariants: - preserve the photographic panel faithfully; - add, remove, duplicate, recolor, reshape, or relocate nothing; - preserve people, faces, age, expressions, hands, clothing, poses, and relationships; - preserve signs, numbers, architecture, object counts, and defining markings. 4. Choose the composition: - default to a landscape left/right direct splice; - use approximately 55% photograph and 45% paper; - obey an explicit top/bottom or alternate ratio request; - keep the main seam perfectly straight and edge-to-edge. 5. Compose the paper panel: - use clean warm off-white paper with subtle fibers and light scan residue; - keep it gently archival, not heavily yellowed, dirty, cracked, burned, or antique; - reserve about 70% as quiet blank paper; - keep the seal-and-caption group at about 30% of the paper panel; - place the group in a balancing corner, never centered unless explicitly requested. 6. Choose a subject-specific seal shape: - circle or oval for compact rounded subjects, faces, icons, or ceremonial motifs; - square frame for signage, dense architecture, storefronts, or when explicitly requested; - panoramic ridge for rows, processions, coastlines, or wide landscapes; - arch for windows, doors, portals, and vaulted subjects; - irregular silhouette for distinctive buildings, vehicles, trees, statues, or objects. 7. Render the seal with dry ink, worn halftone, broken edges, uneven pressure, and slight registration error. Preserve the source's hue relationships with one dominant localized spot color and restrained supporting inks. 8. Add a small faded typewriter caption near, never over, the seal: - title: 1–3 uppercase English words; - subtitle: 2–4 lowercase image words separated by ` / `. 9. Use the built-in image generation/editing tool. Issue one call per source photo. For revisions, edit the latest accepted composite and change only the requested property. 10. Inspect every result. Verify source fidelity, subject count, seal shape, border completeness, scale, corner, straight seam, paper age, text spelling, color memory, and absence of invented content. ## Seal Rules - Make the seal visually compressed rather than a miniature pasted photograph. - Let ink loss and halftone erosion affect the print, not the main panel seam. - For a square seal, show a complete unmistakably square border on all four sides. Allow slight broken ink, worn corners, uneven pressure, and misregistration; never use a clean digital frame or photo mat. - For a circular seal, keep the contour visibly circular while allowing restrained dry-ink gaps. - For a custom silhouette, preserve the subject's defining outline without enclosing it in a generic rectangle. - Keep signs, numbers, faces, and essential structural features legible enough to identify the source. ## Revision Mapping - `方形边框` / `square border`: add one complete worn square ink frame around the existing seal motif; preserve scale and position unless asked otherwise. - `圆形图章` / `circular seal`: recast only the seal within a circular hand-pressed boundary. - `缩小 10%`: scale the seal-and-caption group down by 10%; preserve all other layout values. - `移动到左上/右上/左下/右下`: move only the seal-and-caption group within the paper panel. - `纸张不要那么旧`: reduce yellowing, stains, cracks, and damage; retain subtle warm-white fibers. - `不要那么规整`: roughen the ink boundary with restrained missing ink and halftone erosion; do not create a torn-paper collage. - `保留原来的颜色`: strengthen recognizable source hue relationships while keeping the ink system muted. - `左右拼接` / `上下拼接`: change only the two-panel orientation and reflow the stamp group; keep the boundary straight. ## Guardrails - Produce one flat composite, not a book mockup, scrapbook, poster presentation, or before/after board. - Never use a gradient, dissolve, feather, page turn, gutter, overlap, diagonal split, or decorative separator between the two main panels. - Do not change the source photo to make the seal easier to design. - Do not let the seal dominate the paper panel or eliminate the quiet negative space. - Do not add decorative labels, logos, watermarks, dense copy, or invented objects. - For multiple photos, return one independent finished asset per photo and report every saved path.README.md3.2 KB6c15ac22418bupstream ↗
# Make Photo Stamp Archive **简体中文** · [English](README.en.md) · [日本語](README.ja.md) Make Photo Stamp Archive 是一个 Codex Skill:它可以把一张或多张照片转化为安静、克制的档案拼接作品——一侧忠实保留原照片,另一侧使用暖白纸张与根据主体定制的手工图章,两块画面以笔直边界直接拼接。 调用名称为 `make-photo-stamp-archive`。 ## 核心能力 - 忠实保留原照片中的人物、面孔、物件、建筑、文字、色彩关系与场景逻辑。 - 默认采用横向左右直拼,也可按要求改成上下拼接;主分界始终清晰、笔直。 - 根据主体设计圆形、方框、横向全景、拱形或自定义轮廓图章。 - 呈现干墨、磨损网点、压力不均、边缘缺墨与轻微套色偏差等真实手压质感。 - 支持只修改图章形状、边框、大小、位置、墨色、标题、纸张年代感或拼接方向。 - 多张照片会分别生成独立成品,不合并成拼贴画。 ## 视觉系统 - **照片区:**忠实、真实,默认约占画布的 55%。 - **纸张区:**干净的暖白档案纸,带轻微纤维与扫描痕迹,默认约占画布的 45%。 - **图章组:**体量小、靠角落放置,约占纸张区的 30%。 - **留白:**纸张区约 70% 保持安静、空白。 - **文字:**在图章附近放置小号褪色打字机文字,不覆盖图章。 - **气质:**克制、平面、可触摸、档案化,像一段被保存的记忆。 最终结果始终是一张平面成品,不会做成书本样机、剪贴簿、前后对比板、缩小照片贴片或装饰性海报展示。 ## 作例 | 作例 01 | 作例 02 | | --- | --- | |  |  | | 作例 03 | | --- | |  | ## 运行要求 - Codex,或其他兼容 Skill 的运行环境。 - 能够读取图片,以检查源照片。 - 能够生成或编辑图片,以制作和修改最终成品。 本 Skill 包不包含 API Key、外部字体、脚本或需要额外下载的运行素材。最终画质与源照片保真度取决于宿主环境可用的图像模型。 ## 安装 ```bash git clone https://github.com/Dlcccc71913/skill-make-photo-stamp-archive.git mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills" cp -R skill-make-photo-stamp-archive \ "${CODEX_HOME:-$HOME/.codex}/skills/make-photo-stamp-archive" ``` 如果 Skill 没有立即出现,请重启 Codex。 ## 使用 ```text 使用 $make-photo-stamp-archive,把这张照片制作成带定制图章的档案直拼作品。 ``` 进行局部修改时: ```text 保持照片区完全不变,只把图章缩小 10%,并移动到右上角。 ``` ## 仓库结构 ```text . ├── SKILL.md ├── agents/ │ └── openai.yaml ├── references/ │ └── prompt-template.md ├── README.md ├── README.en.md ├── README.ja.md └── LICENSE ``` ## 许可证 [MIT](LICENSE)references/prompt-template.md4.0 KB9c2e4e2e5ea1upstream ↗
# Prompt Template Use this template for a new composite. Replace bracketed values with facts visible in the source. Remove unused lines. ```text Use case: identity-preserve style-transfer. Asset type: one flat [landscape 4:3 | portrait 3:4] archival stamp composite. Create a [LEFT-AND-RIGHT | TOP-AND-BOTTOM] DIRECT SPLICE. [PHOTO PANEL] — approximately [55%] of the canvas: Preserve the supplied photograph faithfully and realistically. The unique subject is [subject]. Keep [identity, pose, structure, viewpoint, perspective, relative positions, important text, objects, occlusions, environment, lighting, colors, and scene logic]. Crop only as needed to fit the panel. Do not add, remove, duplicate, reshape, recolor, or relocate anything. For people: preserve exact identity, age, facial features, expression, hair, body proportions, hands, fingers, gestures, clothing, and relationships. [PAPER PANEL] — approximately [45%] of the canvas: Use clean warm off-white archival paper with fine visible fibers, subtle cream and pale-gray variation, and light scan residue. Keep it gently archival but not heavily yellowed, dirty, cracked, burned, stained, or distressed. Leave approximately 70% quiet blank paper. Place one compact hand-pressed [CIRCULAR | SQUARE-FRAMED | PANORAMIC | ARCHED | SUBJECT-SILHOUETTE] seal in the [upper-left | upper-right | lower-left | lower-right] area. The entire seal-and-caption group occupies about 30% of the paper panel. Preserve [defining subject features]. Compress the source into a graphic print using rough screen-print, dry ink, worn halftone, uneven pressure, restrained ink loss, and slight registration error. Preserve the original color memory: [main hues]. Use [spot color] as the dominant localized ink with [supporting inks] kept muted. It must look pressed directly into the paper, never pasted as a miniature photograph. [For square seals only: Add one complete unmistakably square ink border around the entire motif. Show all four sides. Use worn corners and slight broken ink, but keep the square boundary readable. No clean digital frame, photo mat, pasted card, or generic rectangle.] DIRECT SPLICE RULE: The two main rectangular panels meet edge-to-edge at one perfectly straight [vertical | horizontal] seam. No gradient, dissolve, feathering, overlap, ink bleed across the seam, page turn, gutter, book spread, diagonal split, mockup, or decorative separator. Add small faded typewriter text near, never over, the seal. Spell exactly: [TITLE] [word / word / word] Mood: quiet, restrained, graphic, tactile, and memory-like. Negative constraints: no changed source subject, no added or removed objects, no changed people, no fantasy scenery, no pasted-photo look, no glossy finish, no clean vector graphics, no decorative sticker, no scrapbook collage, no modern typography, no heavy paper aging, no dense text, no watermark, no page-turn effect. ``` ## Targeted Revision Template ```text Edit the supplied accepted composite. Change only [requested seal, border, scale, position, ink, caption, paper-age, or splice property]. Requested change: [precise instruction]. Keep invariant: the entire photographic panel, source crop, main seam, canvas ratio, paper texture, subject identity and count, scene geometry, original color relationships, negative-space ratio, caption spelling, and every unmentioned detail. Do not regenerate the concept, change the source-photo region, introduce content, cross the seam, add a gradient, or create a page-turn effect. ``` ## Validation Checklist - One faithful photo panel and one paper panel only. - Straight direct seam with no transition effect. - Clean warm-white paper, not excessively old. - Seal-and-caption group about 30%; blank paper about 70%. - Seal sits in a corner and uses a subject-specific shape. - Square seals show a complete readable four-sided ink border. - Source subject, count, structure, text, people, and scene logic remain recognizable. - Original hue memory remains visible through restrained ink. - Caption spelling is exact and small. - No invented content or watermark.agents/openai.yaml251 Bfe6fa3f23b88upstream ↗
interface: display_name: "图章" short_description: "Turn photos into direct-splice archival stamp artworks" default_prompt: "Use $make-photo-stamp-archive to turn this photo into a clean archival paper composition with a compact custom seal."LICENSE1.0 KBb0e01f460c6fupstream ↗
MIT License Copyright (c) 2026 Dlcccc71913 Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.README.en.md3.4 KBb77c1efca543upstream ↗
# Make Photo Stamp Archive [简体中文](README.md) · **English** · [日本語](README.ja.md) Make Photo Stamp Archive is a Codex Skill that turns one or more supplied photos into quiet archival composites: a faithfully preserved photograph joined directly to a warm-white paper panel with a compact, subject-specific hand-pressed seal. The callable skill name is `make-photo-stamp-archive`. ## Core Capabilities - Preserves the source photograph, including people, faces, objects, architecture, text, color relationships, and scene logic. - Uses a clean, straight direct splice—landscape left/right by default, with top/bottom layouts available on request. - Designs circular, square-framed, panoramic, arched, or custom-silhouette seals around the source subject. - Renders tactile dry ink, worn halftone, uneven pressure, broken edges, and restrained registration error. - Supports focused revisions to seal shape, border, scale, position, ink color, caption, paper age, and splice orientation. - Produces one independent finished composite for each supplied photo. ## Visual System - **Photo panel:** faithful and realistic, approximately 55% of the default canvas. - **Paper panel:** clean warm off-white stock with subtle fibers and light scan residue, approximately 45% of the canvas. - **Seal group:** compact and corner-positioned, occupying about 30% of the paper panel. - **Negative space:** roughly 70% of the paper panel remains quiet and empty. - **Caption:** small faded typewriter text placed near the seal, never over it. - **Mood:** restrained, graphic, tactile, archival, and memory-like. The result is always one flat composite—not a book mockup, scrapbook, before/after board, pasted miniature photograph, or decorative poster presentation. ## Examples | Example 01 | Example 02 | | --- | --- | |  |  | | Example 03 | | --- | |  | ## Requirements - Codex or another Skill-compatible runtime. - Image-reading capability for inspecting source photos. - Image generation or editing capability for producing and revising the final composite. No API key, external font, script, or additional runtime asset is included in this Skill package. Final image quality and source-photo fidelity depend on the image model available in the host environment. ## Installation ```bash git clone https://github.com/Dlcccc71913/skill-make-photo-stamp-archive.git mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills" cp -R skill-make-photo-stamp-archive \ "${CODEX_HOME:-$HOME/.codex}/skills/make-photo-stamp-archive" ``` Restart Codex if the Skill does not appear immediately. ## Usage ```text Use $make-photo-stamp-archive to turn this photo into a direct-splice archival artwork with a custom seal. ``` For a focused revision: ```text Keep the photo panel unchanged. Make only the seal 10% smaller and move it to the upper-right corner. ``` ## Repository Structure ```text . ├── SKILL.md ├── agents/ │ └── openai.yaml ├── references/ │ └── prompt-template.md ├── README.md ├── README.en.md ├── README.ja.md └── LICENSE ``` ## License [MIT](LICENSE)README.ja.md4.1 KB8b53bc5eb517upstream ↗
# Make Photo Stamp Archive [简体中文](README.md) · [English](README.en.md) · **日本語** Make Photo Stamp Archive は、1 枚または複数の写真を静かで抑制されたアーカイブ作品へ変換する Codex Skill です。元写真を忠実に保った写真パネルと、被写体に合わせて設計した小さな手押し印を置く温白色の紙パネルを、まっすぐな境界で直接つなぎます。 呼び出し名は `make-photo-stamp-archive` です。 ## 主な機能 - 人物、顔、物、建築、文字、色の関係、場面の論理を含め、元写真を忠実に保持します。 - 標準は横長の左右直結レイアウトです。要望に応じて上下配置にも変更でき、主境界は常に明確で直線的に保ちます。 - 被写体に合わせて、円形、正方形の枠、横長、アーチ形、または独自シルエットの印を設計します。 - かすれたインク、摩耗したハーフトーン、圧力むら、欠けた輪郭、控えめな版ずれによって、実際に手で押したような質感を作ります。 - 印の形、枠、サイズ、位置、インク色、キャプション、紙の古さ、接合方向だけを限定して修正できます。 - 複数の写真には、写真ごとに独立した完成画像を生成します。 ## ビジュアルシステム - **写真パネル:**写実的に忠実さを保ち、標準ではキャンバスの約 55% を使用します。 - **紙パネル:**細かな繊維と軽いスキャン跡を持つ清潔な温白色の紙で、約 45% を使用します。 - **印のグループ:**小さくまとめて隅に配置し、紙パネルの約 30% に収めます。 - **余白:**紙パネルのおよそ 70% を静かな空白として残します。 - **キャプション:**小さく褪せたタイプライター文字を印の近くに置き、印には重ねません。 - **雰囲気:**抑制的、平面的、触覚的、アーカイブ的で、保存された記憶のような印象です。 最終結果は常に 1 枚の平面作品です。本のモックアップ、スクラップブック、ビフォー/アフターボード、縮小写真の貼り付け、装飾的なポスター展示にはしません。 ## 作例 | 作例 01 | 作例 02 | | --- | --- | |  |  | | 作例 03 | | --- | |  | ## 動作要件 - Codex、または互換性のある Skill 実行環境。 - 元写真を確認するための画像読み取り機能。 - 完成画像の生成と修正に使う画像生成または編集機能。 この Skill パッケージには、API キー、外部フォント、スクリプト、追加ダウンロードが必要な実行素材は含まれていません。最終的な画質と元写真の保持精度は、実行環境で利用できる画像モデルに依存します。 ## インストール ```bash git clone https://github.com/Dlcccc71913/skill-make-photo-stamp-archive.git mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills" cp -R skill-make-photo-stamp-archive \ "${CODEX_HOME:-$HOME/.codex}/skills/make-photo-stamp-archive" ``` Skill がすぐに表示されない場合は Codex を再起動してください。 ## 使い方 ```text $make-photo-stamp-archive を使って、この写真をカスタム印付きの直結アーカイブ作品にしてください。 ``` 一部分だけを修正する場合: ```text 写真パネルは完全に変えず、印だけを 10% 小さくして右上へ移動してください。 ``` ## リポジトリ構成 ```text . ├── SKILL.md ├── agents/ │ └── openai.yaml ├── references/ │ └── prompt-template.md ├── README.md ├── README.en.md ├── README.ja.md └── LICENSE ``` ## ライセンス [MIT](LICENSE)
Plain-text copy, no login: /skill-bundles/dlcccc71913-skill-make-photo-stamp-archive.md