Visual AISkills
電商提示詞技能庫探索
瀏覽 庫
修復與編輯
Change image format
Essential

Expand image into banner

Extend an image into website 或 social banner

Use this skill when you need a repeatable image workflow, not a one-off prompt guess. It defines the input, output, tool path, and quality gate before generation starts.

輸入
來源圖
輸出
banner image
Best fit

Change image format

ComfyUI inpaintingFLUXGPT Image 2

Practical format-conversion 工作流 用於 web 和 social publishing

Job to be done

建立 banner image 從 來源圖, using a task-first 工作流 instead of a loose 提示詞 dump.

Workflow type

outpainting, upscale

Quality promise

The output must preserve user intent, survive visual inspection, and be ready for the target channel.

Workflow outline

  1. Prepare the source 輸入: 來源圖.
  2. Pick a first tool path: ComfyUI inpainting, FLUX, GPT Image 2.
  3. 生成 或 edit toward the promised deliverable: banner image.
  4. Review the image 用於 drift, broken details, unreadable text, 和 weak composition.
  5. 儲存 the 提示詞, references, 和 final settings that produced the best reusable result.

Copy-ready skill card

Skill: Expand image into banner

場景: 修復與編輯 / Change image format

輸入: 來源圖

輸出: banner image

Tools: ComfyUI inpainting, FLUX, GPT Image 2

Example 輸入: Input: 來源圖

Goal: describe the target channel, audience, style, 和 non-negotiable details.

Constraints: list what must stay unchanged 和 what can be creatively changed.

品質檢查: the result must match the promised 輸出 和 preserve the user intent.

Branch logic

  • If the user has not provided 來源圖, ask 用於 it before generating the final banner image.
  • If a reference image is provided, preserve the subject identity, proportions, 和 important details before changing style 或 scene.
  • If the first result fails 視覺 inspection, fix the smallest broken area first instead of regenerating the whole image.
  • If the first tool path fails, switch to the next listed option: FLUX, GPT Image 2.

When not to use

  • Do not use this as proof that the generated banner image is factually real 或 legally approved.
  • Do not use it when the user needs professional legal, medical, safety, 或 regulated-compliance judgment.

Copy distribution

Take this skill into your AI tool as a prompt, a workflow card, or a portable SKILL.md draft.

Quality gate

  • The final asset clearly matches the requested banner image.
  • The main subject still reflects the original 來源圖.
  • Text, hands, logos, edges, 和 產品細節 are not visibly broken.
  • The result is usable in the target channel 且不 another full regeneration.

Prompt starter

建立 banner image 用於 this task: Extend an image into website 或 social banner

輸入 available: 來源圖

Example 輸入: Input: 來源圖

Goal: describe the target channel, audience, style, 和 non-negotiable details.

Constraints: list what must stay unchanged 和 what can be creatively changed.

Preferred tool path: ComfyUI inpainting, FLUX, GPT Image 2

Keep the result faithful to the 輸入 和 optimize 用於 the change image format scene.

Evidence and maintenance

Source IDs
IMG-018, IMG-023, IMG-026, IMG-049
Priority
Essential
Library track
Starter
Related intent routes

Related workflows by search intent

Use these routes when the next user question shifts from this workflow into background replacement, product repair, ad variants, listing review, or prompt examples.

Prompt examples

More related skills

All skills