ai-image-generation
Generate and Edit Images with RunComfy
Choosing image models and preparing their inputs can slow creative work. This skill provides RunComfy routing guidance, model-specific prompts, and CLI examples for generation and editing.
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 "ai-image-generation" from https://skillstore.io/skills/gencraft-labs-ai-image-generation.md and its manifest at https://skillstore.io/api/skills/gencraft-labs-ai-image-generation/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「ai-image-generation」。 Prepare a landscape product poster with the exact headline NEW COLLECTION. Do not execute it yet.
預期結果:
- Proposed route: GPT Image 2 text-to-image for a typography-focused poster.
- Size: 1536 by 1024. Place NEW COLLECTION at the top with a centered product below.
- Next step: confirm current endpoint fields and approve execution.
正在使用「ai-image-generation」。 Plan an edit of my product reference. Keep its shape and label unchanged, but replace the background with a white studio.
預期結果:
- Proposed route: Nano Banana 2 Edit using the user-provided reference URL.
- Edit prompt: preserve the product shape, label, and position; replace only the background with a white studio.
- Web search remains disabled. Request approval before sending the reference to RunComfy.
正在使用「ai-image-generation」。 Prepare four square concept images of a ceramic mug for a low-cost draft round.
預期結果:
- Proposed route: Nano Banana 2 text-to-image with four outputs, a square aspect ratio, and draft resolution.
- Prompt direction: a ceramic mug on a marble counter with warm morning lighting and minimal styling.
- These are request settings, not completed images. Confirm current pricing and approve execution.
安全審計
中風險All 199 static matches are false positives involving Markdown, public links, placeholder URLs, or documented token storage. One semantic finding identifies overbroad shell-safety guidance requiring explicit escaping before CLI execution. No evidence found of malicious instructions or credential exfiltration in the supplied file; the external CLI implementation was not reviewed.
已確認的安全問題 (1)
風險因素
⚙️ 外部命令 (50)
🌐 網路存取 (50)
📁 檔案系統存取 (5)
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/gencraft-labs-ai-image-generation/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/gencraft-labs-ai-image-generation?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/gencraft-labs-ai-image-generation?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/gencraft-labs-ai-image-generation/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/gencraft-labs-ai-image-generation.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
gencraft-labs. (2026). ai-image-generation security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/gencraft-labs-ai-image-generation/audits/1BibTeX 引用
@techreport{gencraft-labs-gencraft-labs-ai-image-generation-2026,
author = {gencraft-labs},
title = {ai-image-generation security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/gencraft-labs-ai-image-generation/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: "ai-image-generation security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "gencraft-labs"
date-released: "2026-10-07"
url: "https://skillstore.io/skills/gencraft-labs-ai-image-generation/audits/1"
identifiers:
- type: other
value: "skillstore:gencraft-labs-ai-image-generation:audit:1"
description: "Skillstore immutable audit report identifier"
比較變體
24 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
tool-belt-ai-image-generation
2026-09-09
runcomfy-com-ai-image-generation
2026-09-09
inference-shell-ai-image-generation
2026-09-09
infsh-skills-ai-image-generation
2026-09-09
inference-skills-ai-image-generation
2026-09-09
inference-sh-skills-ai-image-generation
2026-09-09
gencraft-labs-ai-image-generation
2026-10-11
genmedia-labs-ai-image-generation
2026-09-09
skills-shell-ai-image-generation
2026-09-09
prime-skills-ai-image-generation
2026-09-09
agentspace-so-ai-image-generation
2026-10-22
halt-catch-fire-ai-image-generation
2026-09-09
inference-sh-ai-image-generation
2026-09-09
inference-sh-9-ai-image-generation
2026-09-09
inference-sh-8-ai-image-generation
2026-09-09
skills-collective-ai-image-generation
2026-09-09
danielgwilson-ai-image-generation
2026-09-09
Skillstore 評分
評分依據 證據置信度: 中你可以打造什麼
Prepare Campaign Posters
Choose typography-focused generation routes and specify exact headlines, placement, and image sizes for campaign drafts.
Draft Storyboard Concepts
Use fast model routes and lower-resolution batches to explore compositions before selecting final scenes.
Refine Product Photography
Prepare reference-based edits that preserve the product while changing backgrounds, lighting, or localized details.
試試這些提示
Prepare a RunComfy request for [subject] in [setting] with [lighting]. Choose a general-purpose model and save outputs to [directory]. Confirm before execution.
Prepare a landscape poster request with the exact headline [headline]. Specify [font style], [placement], and [colors]. Choose a typography-focused route. Confirm before execution.
Use my reference URL [URL]. Preserve [subject details] and change only [requested change]. Choose the documented edit endpoint. Keep web search disabled. Confirm before execution.
Plan [count] concepts within [budget]. Specify draft resolution, final route, and same-brand refinement. Identify required schema checks. Escape shell arguments safely. Confirm before paid requests.
最佳實務
- Check current endpoint schemas and pricing before execution, then approve the model, output count, and resolution.
- Use only approved reference images and keep web search disabled unless explicitly requested.
- Serialize inputs safely and escape shell arguments consistently; never place tokens in prompts, logs, or shared examples.
避免
- Submitting confidential references to the external service without authorization.
- Assuming JSON encoding alone makes interpolated shell commands safe.
- Requesting large batches or maximum resolution before checking service costs.
常見問題
Which assistants can use this skill?
Does this skill run image models locally?
Do I need an account or paid credits?
Can it edit existing images?
Are exact text and identity preservation guaranteed?
What did the security review establish?
開發者詳細資訊
授權
MIT
Skillstore 修訂
r1
版本提示
作者未聲明版本。
引用
e406bf64b5db6e514879399e87b346fa7ff7c2fb
維護新鮮度
2026/10/11
使用量
0 次下載 · 0 次瀏覽
檔案結構
📄 SKILL.md