screenshotaf

A fully offline, Node.js CLI tool to automatically capture full-page screenshots of web projects. It requires no external APIs, sends no telemetry, and runs 100% locally.

Features

• Project Detection: Automatically identifies whether your project is a single-page app, static HTML site, or requires a dev server (like React, Next.js, or Vite).
• Auto-Discovery: Crawls your running application starting from the homepage to find all reachable routes automatically.
• Scroll-Capture Loop: Instead of one long stitched image, it captures pages into sequential screenshot files corresponding to each viewport height.
• Multi-Device Capture: Captures each page across Mobile (390x844), Tablet (768x1024), and Laptop (1440x900) views at 2x retina pixel density.
• Load Readiness: Guarantees web fonts, lazy-loaded images, and entrance animations have settled before capturing screenshots.
• Clean Execution: Automatically starts your project's dev server and safely shuts it down when finished.

Usage

1. Drop the screenshotaf directory wherever you'd like.
2. To run the tool on a project:
   • Windows: Open the terminal in screenshotaf folder and run run.bat. It takes the current directory as the project target.
   • Mac/Linux: Open a terminal and run:

     npm install
     node index.js /path/to/your/project

Output Structure

The output will be placed in the directory where you ran the tool, named with a timestamp.

myapp-2026-07-08_14-32-10/
  home/
    mobile/
      page-1.png
      page-2.png
    tablet/
      page-1.png
    laptop/
      page-1.png
      page-2.png
      page-3.png
  about/
    mobile/
      page-1.png
    tablet/
      page-1.png
    laptop/
      page-1.png

Known Limitations

• The crawling feature will only discover pages that are explicitly linked within the application using an <a> tag.
• Fully dynamic routes or pages without visible entry points (e.g., /admin or /dashboard/hidden-id) will not be captured automatically. This is expected behavior.