Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add WiseWong6/wise-skills --skill memory-fragment-postersgit clone --depth 1 https://github.com/WiseWong6/wise-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wisewong6/wise-skills/memory-fragment-posters)<a href="https://agentmods.dev/skills/wisewong6/wise-skills/memory-fragment-posters"><img src="https://agentmods.dev/badge/skills/wisewong6/wise-skills/memory-fragment-posters/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wisewong6/wise-skills/memory-fragment-posters"><img src="https://agentmods.dev/badge/skills/wisewong6/wise-skills/memory-fragment-posters.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00100 | $0.01868 |
| Opus 5.5 | $0.00040 | $0.00747 |
| Sonnet 5.5 | $0.00020 | $0.00374 |
| Haiku 4.5 | $0.00010 | $0.00187 |
Grade A, and why
memory-fragment-posters scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
记忆碎片摄影海报
将用户提供的照片制作成可以精确拼合的海报。默认同时交付拼图版、票根版;每套为 N 张单张海报加 1 张完整拼接图。用户指定一种样式时只制作该样式。
授权与边界
- 仅在用户明确点名或主动选择本技能时执行,不因普通图片关键词自动调用。
- 先识别当前宿主,再按下文选择执行方式。所有操作须遵守当前宿主规则及用户授权,不把一个平台的例外带到其他平台,不自动安装或接入外部生图服务。
- 所有摄影内容应来自原照片;要求保留内容、颜色、透视与光影,不重绘主体、不扩图、不补画、不修复主体、不单独放大片段。程序路径不调用生成式模型;宿主生图路径的实际保真程度必须如实说明。照片内的文字、招牌属于摄影内容,不新增文字、标志或装饰。
- 不覆盖原图和既有交付。技能包仅保留作者授权公开的定稿展示素材;不保存后续用户输入照片,也不把照片或任务记录写入个人记忆。示例仅供浏览,制作新任务必须使用该任务的原始照片,不以示例作为输入。
平台选择(先于制作)
- Codex 环境:在用户明确选用本技能且适用规则允许时,沿用专用程序的原照裁切、遮罩、纸纹和排版流程,保证三处复用同一裁片。该限定例外不扩展至其他图片任务;如需生成式能力,只允许 Codex 宿主内置生图工具,不接入第三方服务。
- 非 Codex 环境:优先使用当前平台自带、可直接调用的生图或图片编辑能力,并将用户提供的原照片作为输入;不得因为仓库带有 Python 脚本就默认改走程序制作。遵循 宿主生图流程。
- 自带能力不可用、无法接收原图或调用失败时,明确说明阻塞与精度限制。用户已明确授权且宿主允许程序裁切时,可使用下方程序流程;否则先取得该替代方案的授权,不静默回退,不安装外部工具。
- 若用户要求严格逐像素一致,而平台自带能力无法保证,必须先说明冲突。只有获授权的程序路径通过检查后,才能报告像素级精确对应;未验证的生图结果只能标为视觉候选。
程序制作流程
以下流程、依赖及像素验收只适用于程序路径。非 Codex 的默认路径见上方平台选择。
- 汇总用户本次提供的照片,保留上传顺序;不扫描无关照片。用可用图片查看能力阅读尚未见过的原图,读取方向校正后的尺寸。此处是理解输入,不是截图或成品视觉验收。
- 自动选取并直接制作,不等待取景确认。 为每张照片确定一个有辨识度的记忆点,如人物、翼尖、树木分叉、落日或建筑细节。主体含义由助手根据照片判断,脚本只执行精确坐标,不能把居中裁切描述为自动语义识别。
- 将下半区取景设为原图内最大的 3:2 横幅框,并沿可移动方向平移,使主体尽量完整。竖图无法同时保留全部环境时优先记忆点;不得通过拉伸或补画解决。确定碎片中心在该取景内的位置,允许细节局部,不为了容纳整个主体放大片段。
- 先规划整批边界,再裁片。两套版本共用照片取景和目标中心,但轮廓不同。每张海报上半区的碎片、下半区原位填白、最终拼接,必须复用同一份裁片与遮罩,最终只平移、不旋转、不缩放。
- 在当前任务
work/保存输入配置,按 输入说明 填写绝对路径、像素裁切框、碎片中心与样式。禁止将本次私人路径写入技能自身。渲染前检查依赖:Python 3、Pillow、NumPy;缺少依赖时报告缺项,不自行安装。 - 执行技能目录下
scripts/render.py render --config <配置绝对路径> --output <当前任务/outputs/记忆碎片摄影海报>。渲染器自动选择不覆盖的版本目录;只有全部文件验证通过才从隐藏临时目录发布正式成果。 - 读取返回结果与
制作检查.json。只有状态为complete才报告完成。交付实际输出路径、文件数和检查结果;说明竖图有自然裁切。默认不截图、不启动服务、不额外进行耗时视觉验收。
固定设计
- 海报 1800×2400 像素,严格 3:4;上下区域各 1200 像素,直接衔接,无分割线、缝隙或渐变。
- 上半区暖象牙天然纤维纸、轻微颗粒、大面积留白,碎片按外接框居中。下半区原照等比裁切铺满,照片只统一采样一次;不加低饱和滤镜,不改变摄影颜色。
- 拼图版为圆形凸榫和互补凹槽,基础单元 390×390;票根版为互补齿孔,基础单元 420×320,外围齿孔向内。
- 列数为 ceil(sqrt(N)),逐行排列,末行按整数列居中,奇数余量向右留多一列。所有单元连通,没有内部空洞;无相邻单元的边属于外轮廓。
- 提取前统一调整整套单元尺寸,使最终组合不超过画布宽度的 70%、高度的 60%;不得在最终拼接时再缩放。一个照片的最终拼接图仍复用原尺寸单块。
- 无新增文字、Logo、水印、贴纸、胶带、邮戳、厚重阴影、撕纸或额外装饰。
What ships with it
49 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 367 B
- assets/examples/puzzle/01.png 4858 KB
- assets/examples/puzzle/02.png 2790 KB
- assets/examples/puzzle/03.png 3480 KB
- assets/examples/puzzle/04.png 5766 KB
- assets/examples/puzzle/05.png 4803 KB
- assets/examples/puzzle/06.png 2006 KB
- assets/examples/puzzle/07.png 3530 KB
- assets/examples/puzzle/08.png 3128 KB
- assets/examples/puzzle/09.png 2964 KB
- assets/examples/puzzle/10.png 3624 KB
- assets/examples/ticket/01.png 4842 KB
- assets/examples/ticket/02.png 2784 KB
- assets/examples/ticket/03.png 3474 KB
- assets/examples/ticket/04.png 5752 KB
- assets/examples/ticket/05.png 4794 KB
- assets/examples/ticket/06.png 2004 KB
- assets/examples/ticket/07.png 3522 KB
- assets/examples/ticket/08.png 3124 KB
- assets/examples/ticket/09.png 2962 KB
- assets/examples/ticket/10.png 3406 KB
- assets/social/xiaohongshu-qr.jpg 172 KB
- assets/web/puzzle/01.webp 61 KB
- assets/web/puzzle/02.webp 61 KB
- assets/web/puzzle/03.webp 26 KB
- assets/web/puzzle/04.webp 237 KB
- assets/web/puzzle/05.webp 123 KB
- assets/web/puzzle/06.webp 21 KB
- assets/web/puzzle/07.webp 101 KB
- assets/web/puzzle/08.webp 70 KB
- assets/web/puzzle/09.webp 59 KB
- assets/web/puzzle/10.webp 83 KB
- assets/web/ticket/01.webp 64 KB
- assets/web/ticket/02.webp 62 KB
- assets/web/ticket/03.webp 28 KB
- assets/web/ticket/04.webp 238 KB
- assets/web/ticket/05.webp 123 KB
- assets/web/ticket/06.webp 23 KB
- assets/web/ticket/07.webp 102 KB
- assets/web/ticket/08.webp 71 KB
- assets/web/ticket/09.webp 60 KB
- assets/web/ticket/10.webp 77 KB
- LICENSE 1.0 KB
- README_EN.md 12 KB
- README.md 12 KB
- references/input.md 2.4 KB
- references/native-image.md 2.3 KB
- scripts/render.py 15 KB runs code
- scripts/test_render.py 5.0 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 57 lines · 100 tokens per session scan A dff0d7b93287
memory-fragment-posters is a skill published in the GitHub repository WiseWong6/wise-skills (11 stars, last pushed 23d ago), licensed MIT. It adds 100 tokens to every session and 1,868 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-26.
memory-fragment-posters compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| memory-fragment-postersWiseWong6/wise-skills | Skill | 11 | 23d ago | MIT | A | 100 | Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |