Skip to content
Back to search

mcp-screenshot-website-fast

Quickly screenshots webpages and converts to an LLM friendly size

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/just-every--mcp-screenshot-website-fast/badge.svg)](/listings/just-every--mcp-screenshot-website-fast)

HTML

<a href="/listings/just-every--mcp-screenshot-website-fast"><img src="/api/listings/just-every--mcp-screenshot-website-fast/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Claude Code

claude mcp add screenshot-website-fast -s user -- npx -y @just-every/mcp-screenshot-website-fast

VS Code

code --add-mcp '{"name":"screenshot-website-fast","command":"npx","args":["-y","@just-every/mcp-screenshot-website-fast"]}'

Cursor

cursor://anysphere.cursor-deeplink/mcp/install?name=screenshot-website-fast&config=eyJzY3JlZW5zaG90LXdlYnNpdGUtZmFzdCI6eyJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkBqdXN0LWV2ZXJ5L21jcC1zY3JlZW5zaG90LXdlYnNpdGUtZmFzdCJdfX0=

JetBrains IDEs

Settings → Tools → AI Assistant → Model Context Protocol (MCP) → Add

Choose "As JSON" and paste:

{"command":"npx","args":["-y","@just-every/mcp-screenshot-website-fast"]}

Raw JSON (works in any MCP client)

{
  "mcpServers": {
    "screenshot-website-fast": {
      "command": "npx",
      "args": ["-y", "@just-every/mcp-screenshot-website-fast"]
    }
  }
}

Drop this into your client's mcp.json (e.g. .vscode/mcp.json, ~/.cursor/mcp.json, or .mcp.json for Claude).

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 299 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # @just-every/mcp-screenshot-website-fast
  2. Fast, efficient screenshot capture of web pages - optimized for CLI coding tools. Automatically tiles full pages into 1072x1072 chunks for optimal processing.
  3. <a href="https://glama.ai/mcp/servers/@just-every/mcp-screenshot-website-fast">
  4. <img width="380" height="200" src="https://glama.ai/mcp/servers/@just-every/mcp-screenshot-website-fast/badge" alt="Screenshot Website Fast MCP server" />
  5. </a>
  6. [![npm version](https://badge.fury.io/js/@just-every%2Fmcp-screenshot-website-fast.svg)](https://www.npmjs.com/package/@just-every/mcp-screenshot-website-fast)
  7. [![GitHub Actions](https://github.com/just-every/mcp-screenshot-website-fast/workflows/Release/badge.svg)](https://github.com/just-every/mcp-screenshot-website-fast/actions)
  8. ## Overview
  9. Built specifically for AI vision workflows, this tool captures high-quality screenshots with automatic resolution limiting and tiling for optimal processing by Claude Vision API and other AI models. It ensures screenshots are perfectly sized at 1072x1072 pixels (1.15 megapixels) for maximum compatibility.
  10. ## Features
  11. - 📸 **Fast screenshot capture** using Puppeteer headless browser
  12. - 🎯 **Claude Vision optimized** with automatic resolution limiting (1072x1072 for optimal 1.15 megapixels)
  13. - 🔲 **Automatic tiling** - Full pages are automatically split into 1072x1072 tiles
  14. - 🎬 **Screencast capture** - Record series of screenshots over time with configurable intervals
  15. - 🔄 **Always fresh content** - No caching ensures up-to-date screenshots
  16. - 📱 **Configurable viewports** for responsive testing
  17. - ⏱️ **Wait strategies** for dynamic content (networkidle, custom delays)
  18. - 📄 **Full page capture** by default for complete page screenshots
  19. - 🎥 **Animated WebP export** - Save screencasts as high-quality animated WebP files
  20. - 💉 **JavaScript injection** - Execute custom JS before screencast capture
  21. - 📦 **Minimal dependencies** for fast npm installs
  22. - 🔌 **MCP integration** for seamless AI workflows
  23. - 🪟 **Windows-compatible launcher** for npm-installed MCP usage
  24. - 🔋 **Resource efficient** - Automatic browser cleanup after 60 seconds of inactivity
  25. - 🧹 **Memory management** - Pages are closed after each screenshot to prevent leaks
  26. ## Installation
  27. ### Claude Code
  28. ```bash
  29. claude mcp add screenshot-website-fast -s user -- npx -y @just-every/mcp-screenshot-website-fast
  30. ```
  31. ### VS Code
  32. ```bash
  33. code --add-mcp '{"name":"screenshot-website-fast","command":"npx","args":["-y","@just-every/mcp-screenshot-website-fast"]}'
  34. ```
  35. ### Cursor
  36. ```bash
  37. cursor://anysphere.cursor-deeplink/mcp/install?name=screenshot-website-fast&config=eyJzY3JlZW5zaG90LXdlYnNpdGUtZmFzdCI6eyJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkBqdXN0LWV2ZXJ5L21jcC1zY3JlZW5zaG90LXdlYnNpdGUtZmFzdCJdfX0=
  38. ```
  39. ### JetBrains IDEs
  40. Settings → Tools → AI Assistant → Model Context Protocol (MCP) → Add
  41. Choose "As JSON" and paste:
  42. ```json
  43. {"command":"npx","args":["-y","@just-every/mcp-screenshot-website-fast"]}
  44. ```
  45. ### Raw JSON (works in any MCP client)
  46. ```json
  47. {
  48. "mcpServers": {
  49. "screenshot-website-fast": {
  50. "command": "npx",
  51. "args": ["-y", "@just-every/mcp-screenshot-website-fast"]
  52. }
  53. }
  54. }
  55. ```
  56. Drop this into your client's mcp.json (e.g. .vscode/mcp.json, ~/.cursor/mcp.json, or .mcp.json for Claude).
  57. ## Prerequisites
  58. - Node.js 20.x or higher
  59. - npm or npx
  60. - Chrome/Chromium (automatically downloaded by Puppeteer)
  61. ## Quick Start
  62. ### MCP Server Usage
  63. Once installed in your IDE, the following tools are available:
  64. #### Available Tools
  65. - `take_screenshot` - Captures a high-quality screenshot of a webpage
  66. - Parameters:
  67. - `url` (required): The HTTP/HTTPS URL to capture
  68. - `width` (optional): Viewport width in pixels (max 1072, default: 1072)
  69. - `height` (optional): Viewport height in pixels (max 1072, default: 1072)
  70. - `fullPage` (optional): Capture full page screenshot with tiling (default: true)
  71. - `waitUntil` (optional): Wait until event: load, domcontentloaded, networkidle0, networkidle2 (default: domcontentloaded)
  72. - `waitFor` (optional): Additional wait time in milliseconds
  73. - `directory` (optional): Directory to save screenshots - returns file paths instead of base64 images
  74. - `capture_selector` - Captures a screenshot of a specific DOM element matched by a CSS selector
  75. - Parameters:
  76. - `url` (required): The HTTP/HTTPS URL to capture
  77. - `selector` (required): CSS selector for the element to capture
  78. - `width` (optional): Viewport width in pixels (max 1072, default: 1072)
  79. - `height` (optional): Viewport height in pixels (max 1072, default: 1072)
  80. - `waitUntil` (optional): Wait until event: load, domcontentloaded, networkidle0, networkidle2 (default: domcontentloaded)
  81. - `waitForMS` (optional): Additional wait time in milliseconds
  82. - `selectorTimeoutMS` (optional): How long to wait for the selector to appear before failing (default: 5000)
  83. #### Usage Examples
  84. **Default usage (returns base64 images):**
  85. ```
  86. take_screenshot(url="https://example.com")
  87. ```
  88. **Save to directory (returns file paths):**
  89. ```
  90. take_screenshot(url="https://example.com", directory="/path/to/screenshots")
  91. ```
  92. **Capture a specific element:**
  93. ```
  94. capture_selector(url="https://example.com", selector="#main")
  95. ```
  96. When using the `directory` parameter:
  97. - Screenshots are saved as PNG files with timestamps
  98. - File paths are returned instead of base64 data
  99. - For tiled screenshots, each tile is saved as a separate file
  100. - Directory is created automatically if it doesn't exist
  101. ### take_screencast
  102. Captures a series of screenshots over time to create a screencast. Only captures the top tile (1072x1072) of the viewport.
  103. #### Parameters
  104. - `url` (required): The URL to capture
  105. - `duration` (optional): Total duration in seconds (default: 10)
  106. - `interval` (optional): Interval between screenshots in seconds (default: 2)
  107. - `jsEvaluate` (optional): JavaScript code to execute at the start
  108. - `waitUntil` (optional): Wait strategy: 'load', 'domcontentloaded', 'networkidle0', 'networkidle2'
  109. - `waitForMS` (optional): Additional wait time before starting
  110. - `directory` (optional): Save as animated WebP to directory (captures every 1 second)
  111. #### Usage Examples
  112. **Basic screencast (5 frames over 10 seconds):**
  113. ```
  114. take_screencast(url="https://example.com")
  115. ```
  116. **Custom timing:**
  117. ```
  118. take_screencast(url="https://example.com", duration=15, interval=3)
  119. ```
  120. **With JavaScript execution:**
  121. ```
  122. take_screencast(
  123. url="https://example.com",
  124. jsEvaluate="document.body.style.backgroundColor = 'red';"
  125. )
  126. ```
  127. **Save as animated WebP:**
  128. ```
  129. take_screencast(url="https://example.com", directory="/path/to/output")
  130. ```
  131. When using the `directory` parameter:
  132. - An animated WebP is created with 1-second intervals
  133. - Individual frames are also saved as PNG files
  134. - The animation loops forever by default
  135. - WebP provides excellent quality:
  136. - Full color support (no 256 color limitation)
  137. - Efficient compression for web animations
  138. - Perfect for gradient backgrounds and smooth animations
  139. - Smaller file sizes compared to GIF with better quality
  140. ## Development Usage
  141. ### Install
  142. ```bash
  143. npm install
  144. npm run build
  145. ```
  146. ### Capture screenshot
  147. ```bash
  148. # Full page with automatic tiling (default)
  149. npm run dev capture https://example.com -o screenshot.png
  150. # Viewport-only screenshot
  151. npm run dev capture https://example.com --no-full-page -o screenshot.png
  152. # Wait for specific conditions
  153. npm run dev capture https://example.com --wait-until networkidle0 --wait-for 2000 -o screenshot.png
  154. ```
  155. ### CLI Options
  156. - `-w, --width <pixels>` - Viewport width (max 1072, default: 1072)
  157. - `-h, --height <pixels>` - Viewport height (max 1072, default: 1072)
  158. - `--no-full-page` - Disable full page capture and tiling
  159. - `--wait-until <event>` - Wait until event: load, domcontentloaded, networkidle0, networkidle2
  160. - `--wait-for <ms>` - Additional wait time in milliseconds
  161. - `-o, --output <path>` - Output file path (required for tiled output)
  162. ## Auto-Restart Feature
  163. The MCP server includes automatic restart capability by default for improved reliability:
  164. - Automatically restarts the server if it crashes
  165. - Handles unhandled exceptions and promise rejections
  166. - Implements exponential backoff (max 10 attempts in 1 minute)
  167. - Logs all restart attempts for monitoring
  168. - Gracefully handles shutdown signals (SIGINT, SIGTERM)
  169. For development/debugging without auto-restart:
  170. ```bash
  171. # Run directly without restart wrapper
  172. npm run serve:dev
  173. ```
  174. ## Architecture
  175. ```
  176. mcp-screenshot-website-fast/
  177. ├── src/
  178. │ ├── internal/ # Core screenshot capture logic
  179. │ ├── utils/ # Logger and utilities
  180. │ ├── index.ts # CLI entry point
  181. │ ├── serve.ts # MCP server entry point
  182. │ └── serve-restart.ts # Auto-restart wrapper
  183. ```
  184. ## Development
  185. ```bash
  186. # Run in development mode
  187. npm run dev capture https://example.com -o screenshot.png
  188. # Build for production
  189. npm run build
  190. # Run tests
  191. npm test
  192. # Type checking
  193. npm run typecheck
  194. # Linting
  195. npm run lint
  196. ```
  197. ## Why This Tool?
  198. Built specifically for AI vision workflows:
  199. 1. **Optimized for Claude Vision API** - Automatic resolution limiting to 1072x1072 pixels (1.15 megapixels)
  200. 2. **Automatic tiling** - Full pages split into perfect chunks for AI processing
  201. 3. **Always fresh** - No caching ensures you get the latest content
  202. 4. **MCP native** - First-class integration with AI development tools
  203. 5. **Simple API** - Clean, straightforward interface for capturing screenshots
  204. ## Contributing
  205. Contributions are welcome! Please:
  206. 1. Fork the repository
  207. 2. Create a feature branch
  208. 3. Add tests for new functionality
  209. 4. Submit a pull request
  210. ## Troubleshooting
  211. ### Puppeteer Issues
  212. - Ensure Chrome/Chromium can be downloaded
  213. - Check firewall settings
  214. - Try setting `PUPPETEER_SKIP_CHROMIUM_DOWNLOAD=true` and provide custom executable
  215. ### Screenshot Quality
  216. - Adjust viewport dimensions
  217. - Use appropriate wait strategies
  218. - Check if site requires authentication
  219. ### Timeout Errors
  220. - Increase wait time with `--wait-for` flag
  221. - Use different `--wait-until` strategies
  222. - Check if site is accessible
  223. ## License
  224. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.