imagegen
Create and Edit Project Images
Producing consistent raster assets requires precise prompts, careful edits, and reliable output handling. This skill structures requests, creates or edits visuals, validates results, and saves project-ready files.
使用我的智慧代理安裝
將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。
Review the Skillstore skill "imagegen" from https://skillstore.io/skills/imagegen.md and its manifest at https://skillstore.io/api/skills/imagegen/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。
Agent 可讀資源
當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。
測試它
正在使用「imagegen」。 Create a wide website hero showing a ceramic mug with soft studio lighting and clear space for copy.
預期結果:
A wide product image with a centered ceramic mug, neutral studio background, soft shadows, and open space for page text. Saved as assets/mug-hero.png.
正在使用「imagegen」。 Replace only the background of the supplied chair photo with a bright modern office.
預期結果:
The chair remains unchanged while the background becomes a bright office with matched perspective and lighting. Saved as assets/chair-office.png.
正在使用「imagegen」。 Create three consistent forest item sprites with transparent backgrounds.
預期結果:
- A moss-covered potion bottle with a clean transparent edge.
- A carved wooden key in the same lighting and scale.
- A small leaf-wrapped compass using the shared visual style.
安全審計
低風險Two low-severity static findings are confirmed because the CLI checks whether OPENAI_API_KEY is present; it never prints the key value. Other static findings are false positives from documentation, SVG coordinates, image decoding, dictionary methods, and non-executed backticks. The optional CLI sends selected prompts and images to OpenAI and writes outputs locally after explicit user opt-in.
已確認的安全問題 (1)
能力複核項目 (2)
這些是真實的本機能力，對此技能而言可能是預期行為，因此需要複核，但不會按已確認的惡意行為計分。
風險因素
🌐 網路存取 (9)
📁 檔案系統存取 (7)
🔑 環境變數 (16)
⚙️ 外部命令 (50)
分享與引用此報告
分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。
複製報告連結
https://skillstore.io/skills/imagegen/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/imagegen?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/imagegen?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/imagegen/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/imagegen.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>學術引用 (APA · BibTeX · CFF)
APA 引用
openai. (2026). imagegen security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/imagegen/audits/1BibTeX 引用
@techreport{openai-imagegen-2026,
author = {openai},
title = {imagegen security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/imagegen/audits/1},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "imagegen security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "openai"
date-released: "2026-08-17"
url: "https://skillstore.io/skills/imagegen/audits/1"
identifiers:
- type: other
value: "skillstore:imagegen:audit:1"
description: "Skillstore immutable audit report identifier"
比較變體
2 個可安裝變體每位作者的技能都保留為獨立的可安裝項目。推薦變體依 Skillstore 證據排序。
為何此變體排名第一
Skillstore 評分
評分依據 證據置信度: 中你可以打造什麼
Build Website Visuals
Create hero images, product scenes, banners, and transparent assets sized for a website project.
Produce Game Art
Generate sprites, textures, concept art, and visual variants with shared style constraints.
Prepare Campaign Assets
Create product mockups and advertising visuals with exact copy, audience context, and layout requirements.
試試這些提示
Create a [asset type] showing [subject]. Use [style] and [composition]. Avoid logos, watermarks, and unrequested text.
Edit the provided [image role]. Change only [target change]. Preserve [subject, layout, colors, and details]. Save the result as [filename].
Create a [campaign asset] for [audience]. Show [product] in [scene] with [lighting]. Include the exact text: "[copy]". Keep space for [layout need].
Produce [count] distinct [asset type] images for [project]. Vary [controlled dimension] while preserving [shared constraints]. Validate composition, text, and consistency. Save descriptive filenames in [directory].
最佳實務
- State the asset purpose, subject, style, composition, and constraints before generation.
- For edits, name every element that must remain unchanged.
- Inspect text, edges, dimensions, and project paths before accepting the final image.
避免
- Do not use vague prompts when exact layout, copy, or identity matters.
- Do not replace editable vector assets with generated raster approximations.
- Do not send sensitive source images through the optional CLI without reviewing external data exposure.