# Screenshot Website Fast

> Use this tool when you need to capture high-quality, full-page website screenshots for visual analysis or automated content processing. It solves problems such as generating code from website designs, analyzing web pages visually, and processing large pages with intelligent image tiling. The tool accepts website URLs as input and outputs 1072x1072 pixel image tiles, making it ideal for AI assistants that require fresh, properly-sized screenshots for vision model processing.

Canonical page: https://skillsregistry.net/skills/just-every-screenshot-website-fast  
JSON: https://api.skillsregistry.net/v1/skills/just-every-screenshot-website-fast

## Description

This MCP server provides fast, AI-optimized website screenshot capture using Puppeteer with automatic image tiling for large pages. Built by Just Every using TypeScript, it captures full-page screenshots and intelligently splits them into 1072x1072 pixel tiles optimized for Claude Vision API processing, with robust error handling including browser restart capabilities and retry logic for problematic sites. The implementation offers both CLI and MCP server interfaces with no caching to ensure always-fresh content, making it valuable for AI assistants that need to analyze web pages visually, generate code from website designs, or perform automated web content analysis where high-quality, properly-sized screenshots are essential for vision model processing.

## Trust

- **Trust score (0–1):** 0.82
- **Verification tier:** verified
- **Last scanned:** 2026-09-19

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** browser-automation
- **Updated:** 2026-09-19

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/just-every-screenshot-website-fast)
- **Repository:** <https://github.com/just-every/mcp-screenshot-website-fast>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "just-every-screenshot-website-fast"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/just-every-screenshot-website-fast` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/just-every-screenshot-website-fast/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt