Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/ZJU-REAL/Easelnpx agentmods add skills/zju-real/easel/ecom-details-imageWrote 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/zju-real/easel/ecom-details-image)<a href="https://agentmods.dev/skills/zju-real/easel/ecom-details-image"><img src="https://agentmods.dev/badge/skills/zju-real/easel/ecom-details-image/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/zju-real/easel/ecom-details-image"><img src="https://agentmods.dev/badge/skills/zju-real/easel/ecom-details-image.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 67 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- medium Data Exfiltration · line 55 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 | $0.00101 | $0.02138 |
| Opus 5 | $0.00051 | $0.01069 |
| Sonnet 5 | $0.00020 | $0.00428 |
| Haiku 4.5 | $0.00010 | $0.00214 |
Grade A, and why
ecom-details-image 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 11d 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 — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ecom-details-image Skill
当用户需要视觉策略、图片 Prompt、商品主图、营销图、社媒图、广告图、电商 PDP 视觉，或要求直接 AI 生图时，使用这个 Skill。
两种模式：
- Brief / Prompt 模式：只输出视觉简报和可执行图片 Prompt。
- Generate 模式：当用户明确要求"生图、生成图片、出图、render image"时，先输出最终 Prompt，再调用
scripts/generate_image.py。
审美方向对齐 card-design：详情页信息图 / 营销图的视觉方向遵守同一套去 AI 廉价感原则——禁蓝紫科技渐变、禁 emoji 当图标、字大字细、留白克制、填满不空。把这些写进 Prompt 的正向/否定约束里（本 SKILL 出 t2i Prompt，非 HTML 渲染，故是"审美方向"层面对齐，不走 render_card）。
不要暴露、索要、写入、提交或回显真实 API key。使用者必须通过自己的环境变量配置 API。
References（按需加载，不要一次性全读）
| 文件 | 内容 | 何时读 |
|---|---|---|
references/templates.md |
25 个场景模板匹配表、使用方式、风格变体速查 | 判断场景类型、匹配模板时 |
references/templates/*.json |
具体场景模板（prompt_template/variants/category_tips/anti_ai_tips） |
只读匹配到的那一个 |
references/image-gen-rules.md |
通用 Prompt 结构、文生图铁律、精简原则、Anti-AI 技巧、翻车点防护 | 写每一条 Prompt 时 |
references/campaign-style-lock.md |
多图任务的 Campaign Style Lock 规则与默认模板 | 任务含多张图时 |
references/pdp-sequences.md |
转化驱动诊断、主图/详情页序列、多角度镜头、详情页信息图结构、字体搭配 | 商品图/详情页/PDP 任务时 |
核心流程
- 判断视觉任务类型和场景 → 读
references/templates.md匹配模板。 - 从
references/templates/读取匹配到的那一个 JSON，取prompt_template、variants、category_tips作为 Prompt 基础结构。 - 只收集会实质影响图片结果的缺失信息（见下方最小输入）。
- 构建视觉简报。
- 多图任务：先按
references/campaign-style-lock.md建立 Campaign Style Lock，锁定整套图的色板、冷暖调、字体、背景、光线、布局和图标风格。 - 按
references/image-gen-rules.md写出可执行图片 Prompt（保持简洁，逐条对照铁律）；多图任务必须把同一段 Campaign Style Lock 原样放进每张 Prompt。 - 商品图/详情页/营销图：按
references/pdp-sequences.md先做转化驱动力诊断，再排序列。 - 用户要求电商详情页 / PDP / 主图堆栈 / 整套商品图时，默认输出 5 张主图 + 7-9 张详情页图片 的图片包（详情页每屏必须是电商信息图格式，见
references/pdp-sequences.md）。 - 用户要求直接出图 → 调用
scripts/generate_image.py；用户提供参考产品图时传入--image。 - 返回 Prompt、生成文件路径和关键假设。
最小输入
任何视觉任务优先确认：目标、用途（主图/广告图/社媒图/Banner/PDP 模块/缩略图等）、主体、受众与语境、风格、构图与比例、是否需要图内文字、负面约束。缺少非关键字段时，明确假设后继续，不要无谓阻塞。
生图脚本调用（scripts/generate_image.py）
直接生图走 apimart.ai 图像生成接口（模型由 IMG_MODEL 指定，model-agnostic，异步轮询）；也可改用统一生图入口 skill ai-image-gen。优先在 .claude/skills/ecom-details-image/ 放 .env，不要把真实 API key 写进仓库：
What ships with it
32 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.
- .env.example 469 B
- EASEL-META.md 1.5 KB
- references/campaign-style-lock.md 2.7 KB
- references/image-gen-rules.md 5.3 KB
- references/pdp-sequences.md 12 KB
- references/templates.md 2.6 KB
- references/templates/01-hero-image.json 2.4 KB
- references/templates/02-lifestyle-scene.json 2.7 KB
- references/templates/03-flat-lay.json 2.6 KB
- references/templates/04-detail-macro.json 2.6 KB
- references/templates/05-poster-banner.json 2.8 KB
- references/templates/06-social-media.json 3.0 KB
- references/templates/07-ugc-style.json 3.9 KB
- references/templates/08-model-showcase.json 3.1 KB
- references/templates/09-before-after.json 2.4 KB
- references/templates/10-packaging.json 2.5 KB
- references/templates/11-infographic.json 2.5 KB
- references/templates/12-creative-concept.json 2.4 KB
- references/templates/13-size-spec.json 2.5 KB
- references/templates/14-multi-product.json 2.4 KB
- references/templates/15-livestream.json 3.3 KB
- references/templates/16-try-on-virtual.json 2.6 KB
- references/templates/17-exploded-view.json 2.7 KB
- references/templates/18-ghost-mannequin.json 2.8 KB
- references/templates/19-multi-angle-grid.json 2.6 KB
- references/templates/20-magazine-editorial.json 3.0 KB
- references/templates/21-seasonal-campaign.json 2.6 KB
- references/templates/22-luxury-atmospherics.json 2.7 KB
- references/templates/23-device-mockup.json 2.8 KB
- references/templates/24-storefront.json 2.7 KB
- references/templates/25-sports-campaign.json 2.7 KB
- scripts/generate_image.py 18 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.
- 11d ago First seen · 112 lines · 101 tokens per session scan A e1e3c6acd890
ecom-details-image is a skill published in the GitHub repository ZJU-REAL/Easel (794 stars, last pushed yesterday), licensed Apache-2.0. It adds 101 tokens to every session and 2,138 once invoked, about $0.0005 per session on Opus 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 2026-08-30.
Other skills, from other repositories
p-image-try-on
Use when someone wants virtual try-on — dress a person in clothes from reference photos for fashion or ecommerce.
gemini-image-generator
Use when generating professional posed product images for e-commerce using Gemini AI with optimized prompts.
byted-livesaas-master
A control tool for managing business livestreams, including rooms, comments, viewers, product cards, coupons, and live-session settings.
byted-ind-ecom-product-video-prompt
A structured prompt-writing guide for creating e-commerce product videos with Seedance 2.0. It turns one or more product images into a product showcase script using scene settings, timed shots, and output constraints.
ppt-generation
Generate PPTX presentations from slide plan + content.
chart-visualization
Generate charts: select type, extract data, render image.