AI Image Upscaling

该工具调用HiResolutionPhotos高性能AI引擎，无需API密钥即可将普通图片升级为清晰的4K/8K画质。支持PNG、JPG、WebP格式（单文件≤10MB），可选择2/4倍缩放。适用于本地/网络图片画质提升、小图转生产级4K图等场景。每小时每IP限15次请求，处理完成后返回安全下载链接，需引导用户自行下载。

setdemos

热度 40.51

最近更新 2026-05-11 17:40:07

分类：创作依赖：读写文件 / 第三方 API应用场景：图像生成

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add hiresolutionphotos-upscaler

Metadata

name: hiresolutionphotos-upscaler
description: AI Image Upscaling API via HiResolutionPhotos. Perfect for ai image upscaling, api image upscaling, and high resolution upscaler tools. Send low-res images and get back incredibly sharp 4K and 8K results natively utilizing dedicated hardware. The final output is provided via a secure private retrieval webpage ensuring privacy and seamless downloading.
version: 1.0.1
metadata: {"openclaw":{"requires":{"bins":["curl"]},"emoji":"📸","homepage":"https://hiresolutionphotos.com"}}

HiResolutionPhotos — AI Image Upscaling

Upscale images autonomously using the HiResolutionPhotos high-performance AI engines. No API key required. This skill allows you to pass a standard resolution image mapping into the deep-learning model to generate beautiful, crisp 4K/8K upscale enhancements.

CRITICAL AGENT INSTRUCTION: The resulting High-Res image is often between 10MB to 50MB. OpenClaw Agents MUST NOT attempt to download the raw binary output. YOU WILL CRASH. Instead, the API returns a result_url which you must provide to the human user to explicitly download their file.

When to use this skill

You need to perform standard ai image upscaling on a user's local file or web image.
A user tasks you with utilizing an api image upscaling pipeline to improve visual quality.
You want to convert a small 512px sketch or photo into a massive, production-ready 4K render.

Rate limits & Constraints

Submission Limit: 15 upscale requests per hour per IP. Exceeding this will return a 429 Error.
Max Input File Size: 10 MB limit on the source image.
Allowed Formats: PNG, JPG, WebP.

1. Submit an Image for Upscaling

Endpoint: POST https://hiresolutionphotos.com/api/upscale Content-Type: multipart/form-data

Required Fields

Field	Description
`image`	The physical source file (PNG, JPG, WebP)
`scale`	Integer: `2` or `4`

Example

curl -s -X POST -F "image=@/path/to/image.jpg" -F "scale=4" https://hiresolutionphotos.com/api/upscale

Response

{
  "id": "abc123xyz",
  "status": "IN_PROGRESS",
  "backend": "local"
}

2. Poll the Status & Retrieve Result

Because AI processing takes time (usually between 10 to 60 seconds depending on scale), you must poll the server. You must append &agent=true to force generation of the safe retrieval website link!

Endpoint: GET https://hiresolutionphotos.com/api/upscale/status?id=<job_id>&agent=true

Example

curl -s "https://hiresolutionphotos.com/api/upscale/status?id=abc123xyz&agent=true"

Example Polling Loop Response

While processing:

{
  "status": "IN_PROGRESS"
}

When finished:

{
  "status": "COMPLETED",
  "result_url": "https://hiresolutionphotos.com/result/abc123xyz"
}

3. The Retrieval Workflow (MANDATORY)

Once you receive the result_url:

DO NOT run curl on the result_url. It is an HTML Next.js webpage, not a raw image, and parsing it will yield nothing!
Simply output the URL back to your human user in markdown:

"Your image has successfully been upscaled! You can securely view and download the High-Resolution 4K format here: View Upscaled Image"