---
name: visual-dx.hugen.tokyo
description: visual-dx.hugen.tokyo provides a single API for capturing screenshots of publicly accessible web pages and returning the result as a base64-encoded JPEG or PNG. It supports configurable viewport dimensions, full-page capture, dark mode rendering, ad blocking, and CSS selector targeting for element-level snapshots.
host: visual-dx.hugen.tokyo
---

# visual-dx.hugen.tokyo

This host is a focused screenshot utility serving agents that need rendered visual output from URLs. It is not a general-purpose browser automation or scraping platform — its sole function is producing image captures of web content with a handful of rendering options. It suits workflows where a visual representation of a page or specific DOM element is required as an artifact or for downstream image processing.

## When to use this host

Use this host when an agent needs a rendered visual snapshot of a webpage or a specific DOM element — for example, to verify layout, document a UI state, or produce an image artifact for downstream processing. Do not use it for PDF generation (use a dedicated HTML-to-PDF API), structured data extraction or scraping (use a scraping API), or real-time video capture. If the goal is to parse text or data from a page rather than capture its visual appearance, route to a scraping or headless-browser data extraction service instead.

## Capabilities

### Webpage Screenshot Capture

Renders any publicly accessible URL into a JPEG or PNG image, with control over viewport size, full-page vs. viewport-only capture, dark mode, ad blocking, and CSS selector targeting for isolated element screenshots.

- **`capture-webpage-screenshot`** — Captures a screenshot of any publicly accessible URL and returns the image as a base64-encoded JPEG or PNG, with configurable viewport, full-page capture, dark mode, and CSS selector targeting.

## Skill reference

### `capture-webpage-screenshot`

**Visual Screenshot API** — Captures a screenshot of any publicly accessible URL and returns the image as a base64-encoded JPEG or PNG, with configurable viewport, full-page capture, dark mode, and CSS selector targeting.

*Use when:* Use when an agent needs a visual snapshot of a webpage or image URL, such as verifying page layout, capturing a specific DOM element, or producing a rendered image of web content for downstream processing.

*Not for:* Do not use for real-time video capture, PDF generation, or scraping structured data from pages — use a dedicated scraping or PDF API instead.

**Inputs:**

- `url` (string, required) — URL to capture. Maximum 4096 characters.
- `width` (integer) — Viewport width in pixels. Range: 320–3840. Default: 1280.
- `height` (integer) — Viewport height in pixels. Range: 240–2160. Default: 720.
- `full_page` (boolean) — When true, captures the full scrollable page rather than just the viewport. Default: false.
- `max_height` (integer) — Maximum height in pixels for full-page capture. Range: 720–32768. Default: 16384.
- `format` (string) — Output image format. Accepts 'jpeg' or 'png'. Default: 'jpeg'.
- `quality` (integer) — JPEG compression quality. Range: 1–100. Default: 80. Only applies when format is 'jpeg'.
- `dark_mode` (boolean) — When true, renders the page using a dark color scheme. Default: false.
- `block_ads` (boolean) — When true, blocks ads and cookie banners before capturing. Default: true.
- `selector` (string) — CSS selector targeting a specific DOM element to capture instead of the full viewport. Maximum 500 characters. Null captures the full page.

**Returns:** Returns a JSON object with the captured URL, viewport dimensions, image format, and the full screenshot as a base64-encoded JPEG string.

**Example:** `GET https://visual-dx.hugen.tokyo/visual/screenshot?url=https%3A%2F%2Fexample.com&width=1280&height=720&format=jpeg&quality=80&full_page=false&dark_mode=false&block_ads=true`

---