old-photo-restoration

---
name: old-photo-restoration
description: "Restore and enhance old photographs — repair scratches/creases/mold/tears, deblur faces, correct fading and yellowing, upscale quality, and naturally colorize black-and-white photos. Use when a supplied photograph is old, damaged, faded, or black-and-white and the goal is faithful recovery. Fidelity first — never swap faces, over-beautify, or alter a person's identity. Do not use for creating new stylized portraits. Do not use for ID-photo specs. Do not use for editing intact modern photos (object removal, background swap, retouch, anime restyle)."
homepage: https://deepkolor.ai
author: DeepKolor
license: MIT
tags: [image-generation, restoration, vintage]
version: lite
---

# Old Photo Restoration

> **Lite edition** — routing and core rules for agent discovery. Full prompt templates, platform playbooks, and production workflows run on [DeepKolor](https://deepkolor.ai).

Restore damaged, faded, or black-and-white photos — repair, deblur, color correct, upscale, natural colorize.

## When to use

See `description` in frontmatter — the agent loads it at startup for routing.

## Workflow

- Diagnose degradation first; fidelity over beautification.
- Preserve original identity and vintage character.
- Pipeline: repair → enhance → colorize (optional), one step at a time if needed.

## Defaults

aspectRatio=auto, resolution=1K

## Tool: `create_image_task`

```ts
{
  prompt: string,
  scene: "text-to-image" | "image-to-image",
  aspectRatio?: string,
  imageUrls?: string[],
  resolution?: "1K"
}
```

Available when your agent runtime exposes DeepKolor generation tools. Otherwise use this skill as a prompt-engineering guide and adapt tool names to your stack.

## Boundaries

- Old/damaged photos only — not modern photo edits or new stylized portraits.
- Never swap faces or over-beautify.

## Output

After generation, reply briefly: what was created, key parameters, and one useful next step. Return a full prompt only when the user explicitly asks.

## Full version

Production-grade references (SOPs, templates, platform rules) ship inside [DeepKolor](https://deepkolor.ai) — not in this public lite package.