amazon-listing-pack-sop SKILL.md skillA

amazon-listing-pack-sop from orieileen/canvex is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add orieileen/canvex/amazon-listing-pack-sop@git:20260910.9031ce8

Pins git:20260910.9031ce8 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
orieileen/canvex
Path
backend/studio/services/agent/skills/amazon-listing-pack-sop/SKILL.md
Versions recorded
2
Latest version
git:20260910.9031ce8, committed Sep 10, 2026
First version recorded
Jun 20, 2026
Size
17,268 bytes
Repository stars
36
License
MIT

用户要一整套 Amazon 上架图时用这条技能。常见说法有「做一套 Amazon 主图」「7 张套图」「Amazon listing 套图」「上架图包」「Amazon 7-pack」「listing image set」。并行生成标准的 7 张套图 —— 主图、信息图、另一角度、细节特写、两张场景图、尺寸对比 —— 七张共用同一张 img2img 源图, 所以每个角度里都是同一个产品。提示词故意留白, 让 img2img 自己决定这个产品该突出哪些细节(杯子就拍手柄, 挂画就拍背面挂钩, 地毯就拍底面纹理)。.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# amazon-listing-pack-sop

> **本文说明文字用中文, 但发给图像模型的 prompt 一律保持英文。**
> 预设生图模型是 gpt-image-2, 英文提示词的出图质量明显更稳 —— 而这条技能的全部价值
> 就在提示词质量上。下面表格里那 7 段、以及示例里 `prompt=` 的内容, 都不要翻译。

## 什么时候用

用户这一轮要的是**一整套互相配套的 Amazon 上架图**时匹配这条技能。常见说法:

- 「做一套 Amazon 主图 / 7 张套图 / listing 套图 / 上架图包」
- 「Amazon 7-pack / listing image set / product photography pack」
- 「出一套上架图」

**不要用在:**
- 单张图的请求 → 走 `image-prompt-sop`
- 没有「套 / 套图 / pack / set」字样的随口一句「画一张图」「来张图」——
  别把七张的流程强加在一个只想要一张的人身上
- 视频请求 → 不相干, 走 `generate_video`

## 前置检查: 必须有附件(硬门槛)

这条技能是图生图。**七个角度共用同一张源图**, 用户真实的产品才能贯穿整套 —— 这正是
「套图」这两个字的意义。没有共用源图的话, 每个角度都变成从一句泛泛的「这个产品」文生图,
结果就是七个互不相干的随机产品。用户抱怨的「牛头不对马嘴」说的就是这个失败模式。

派发任何 `generate_image` 之前, 先确认对话里有一条标着
`[Canvas attachments for this turn]` 的系统消息。**没有就必须拒绝**, 按用户的语言回:

- 中文: 「做这套图需要参考你的实际产品图。请先在 canvas 上选中一张产品图, 点右上
  『Send to chat』把它附上, 然后重新发指令, 我会按 7 个角度生成统一套图。」
- 英文: "I need your actual product as a reference to generate this set.
  Please select a product image on canvas, click 'Send to chat' to attach it,
  then re-send your request and I'll produce the 7-angle pack."

没有附件时**一次 `generate_image` 都不要调**。工具层对套图调用(任何带 `slot_index`
的调用)本来也会硬拒 —— 所以跳过这道前置检查的下场是七个调用全被工具层打回, 报错原文
照样是上面那句话。省掉这一轮往返, 提前拒。
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add orieileen/canvex/amazon-listing-pack-sop@sha256:bd1624e52dd29655

The line at the top of this page pins the label git:20260910.9031ce8, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_tkmbqnpfozyzrxqd.svg)](https://markdownregistry.com/a/art_tkmbqnpfozyzrxqd)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 2 versions of this file, between Jun 20, 2026 and Sep 10, 2026, about one every 82 days.

versioncommittedcommitsizeaudit
git:20260910.9031ce8 latest2026-09-10 9031ce8 17,268 BA view · diff
git:20260620.1e9857f2026-06-20 1e9857f 16,674 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (17268 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

orieileen/canvex · 36 stars · license MIT · pushed 2026-09-22 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_tkmbqnpfozyzrxqd
GET https://markdownregistry.com/api/v1/resolve?ref=orieileen/canvex/amazon-listing-pack-sop
GET https://markdownregistry.com/api/v1/blob/bd1624e52dd2965577e6c13a8684c7753316f2bb3c0ec0b4907ef816128d601f

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from orieileen/canvex

image-prompt-sop SKILL.md skill
orieileen/canvex · backend/studio/services/agent/skills/image-prompt-sop/SKILL.md · 用户要在画布上生成、创建、画一张或可视化一个图像时用这条技能 —— 产品图、场景图、样机、营销视觉、概念图、插画、图标都算。这条技能负责把一句含糊的需求改写成高质量提示词, 并在调用 `generate_image` 之前定好尺寸和张数。
git:20260910.ebe1fa7 · audit A · 36 stars

Every file in orieileen/canvex

Browse by kind, by grade A, or by owner.