Visual AISkills
이커머스프롬프트스킬탐색
수정 및 편집
수정 failures
Essential

고치기 hands 및 faces

수정 common AI portrait failures

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.

입력
broken image
출력
repaired image
Best fit

수정 failures

ADetailerImpact PackGPT Image 2

High-intent repair page

Job to be done

만들기 repaired image 에서 broken image, using a task-first 워크플로 instead of a loose 프롬프트 dump.

Workflow type

inpainting, detailer

Quality promise

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

Workflow outline

  1. Prepare the source 입력: broken image.
  2. Pick a first tool path: ADetailer, Impact Pack, GPT Image 2.
  3. 생성 또는 edit toward the promised deliverable: repaired image.
  4. Review the image 용 drift, broken details, unreadable text, 및 weak composition.
  5. 저장 the 프롬프트, references, 및 final settings that produced the best reusable result.

Copy-ready skill card

Skill: 고치기 hands 및 faces

장면: 수정 및 편집 / 수정 failures

입력: broken image

출력: repaired image

Tools: ADetailer, Impact Pack, GPT Image 2

Example 입력: Product image: generated bottle 비주얼 로 warped label edge

Issue to fix: distorted 라벨 텍스트 및 uneven cap 형태

Must preserve: composition, lighting, product 색상, 소재

Edit scope: repair only the broken areas

Do not change: 주장, logo, flavor, size, 또는 packaging structure

품질 확인: the result must match the promised 출력 및 preserve the user intent.

Branch logic

  • If the user has not provided broken image, ask 용 it before generating the final repaired image.
  • If a reference image is provided, preserve the subject identity, proportions, 및 important details before changing style 또는 scene.
  • If the first result fails 비주얼 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: Impact Pack, GPT Image 2.
  • If the task depends on masks, text, 또는 object edges, inspect those details at full size before accepting the 출력.
  • If the image will be 게시됨 commercially 또는 publicly, run the risk guardrails before export.

When not to use

  • Do not use this as proof that the generated repaired image is factually real 또는 legally approved.
  • Do not use it when the user needs professional legal, medical, safety, 또는 regulated-compliance judgment.
  • Do not publish the result until 권리, consent, authenticity, 및 platform constraints have been checked.

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 repaired image.
  • The main subject still reflects the original broken image.
  • Text, hands, logos, edges, 및 제품 디테일 are not visibly broken.
  • The result is usable in the target channel 없이 another full regeneration.
  • Confirm the user has 권리 또는 consent to use the source photo 및 avoid exposing private personal details.

Prompt starter

만들기 repaired image 용 this task: 수정 common AI portrait failures

입력 available: broken image

Example 입력: Product image: generated bottle 비주얼 로 warped label edge

Issue to fix: distorted 라벨 텍스트 및 uneven cap 형태

Must preserve: composition, lighting, product 색상, 소재

Edit scope: repair only the broken areas

Do not change: 주장, logo, flavor, size, 또는 packaging structure

Preferred tool path: ADetailer, Impact Pack, GPT Image 2

Keep the result faithful to the 입력 및 optimize 용 the 수정 failures scene.

Evidence and maintenance

Source IDs
IMG-018, IMG-024, IMG-025, IMG-055
Priority
Essential
Library track
Starter

More related skills

All skills