True Face Photo Repair · Open Skill 공개 스킬

True Face Photo Repair

True Face Photo Repair은 Digen의 공개 Open Skill입니다. 열어 보면 이 워크플로가 무엇을 만드는지 알 수 있습니다.

@HarperDiaz

Custom

4 1 7

https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4284956/2026-10-16/kk9vx422.webp

# True Face Photo Repair

Sharper, cleaner, still you. Blur, noise, banding and double edges disappear while your face, features and character stay exactly as they are — restoration, not beautification.

## When to use

- The user wants a blurry, noisy or awkwardly edited portrait cleaned up without changing the face (faithful photo restoration, fix blur and double edges, sharpen a face photo naturally)
- A variation on a previous result: strong for very damaged photos

Not for: Video output (use a video skill).

## Inputs and gates

| Input | Required | Default | Gate |
|---|---|---|---|
| Photo | Yes | — | One adult portrait; blur, noise, grain and editing artifacts are expected. Ask for another photo only if no face is visible |
| Strength | No | `natural` | natural (default), strong for very damaged photos |
| Ratio | No | `3:4` | 3:4 (default); match the original on request |

- If the person may be under 18, stop and decline politely. Real-person photos must be of adults the user has the right to use.
- Fashion, dark and horror themes are fine as stills; refuse explicit sexual content and anything sexualizing minors.

## Steps

1. Read `references/image-edit.md` for prompt structure.
2. **Identity read**: `recognize_image(image_url=<photo>, prompt='Describe this adult for identity preservation: face shape, eye shape and color, brows, nose, lips, skin tone, apparent age range, hair color and length, distinctive marks, glasses. One compact line.')`.
3. **Generate**: `generate_image(prompt=<PROMPT>, input_urls=[photo], model='i2i.hd', orientation='portrait', aspect_ratio='3:4', name='True Face Photo Repair', entity_type='skill_generation')`. Merge the user's details and options into the template.
4. **Self-check, one retry**: look at the result. If a person photo was used and the face no longer reads as the same person, regenerate with the identity clause first. If someone glancing at it would not instantly see the intended change (it reads as a light filter, the old clothes or background remain, or the style is faint), regenerate once with the change described bigger and more concretely.
5. Reply in one or two sentences, e.g. "All set. Your true face photo repair is on the canvas. Next I can do strong for very damaged photos." Results appear on the canvas; never paste URLs.

## Prompt templates

**PROMPT**

```text
The same adult from the reference: <identity line>. Same face structure, eyes, skin tone and recognizable likeness; no beautification or face reshaping. Intentional changes: the same image faithfully restored: remove blur, grain, compression noise, color banding, halos, double contours and awkward compositing seams; recover natural detail in eyes, lashes, hair strands and skin pores; keep the exact face, proportions, expression, makeup, skin tone, hairstyle, clothing, framing and background; no beautification, no smoothing into plastic skin, no reshaping, true-to-life color. <user details and options>. Restoration only: crisp natural detail and clean color with zero change to identity, expression, composition or styling. Lighting and materials described concretely; composition suited to portrait 3:4. Photoreal unless the style says otherwise, fully clothed, no text unless requested.
```

Quality bar: A faithful repair, not a beautification: compare the result with the original side by side; if the face shape, features or expression shifted at all, regenerate once with a stronger preserve list. Report what was fixed (blur, noise, seams).

## Failure handling

- `success=false`: say what failed in one sentence; retry once only if transient; never invent links.
- The face changes: regenerate once with the identity clause first; then say so honestly.

## Examples

**Default**: the user says "this one" with a photo. Use strength `natural`, ratio `3:4`, 3:4. Reply: "All set. Your true face photo repair is on the canvas."
**Variation**: "next variation". Keep the same inputs and change one option (strong for very damag