Capture LogoCapture

AI Agent Skill

Install Capture as an Agent Skill for Claude Code. Enable Claude to capture screenshots, generate PDFs, and extract web content using the Capture CLI.

Add Capture as an Agent Skill to give Claude the ability to capture screenshots, generate PDFs, and extract content from any URL.

Prerequisites

  1. Install the Capture CLI
  2. Set environment variables CAPTURE_KEY and CAPTURE_SECRET

Installation

npx skills add techulus/capture-skills

For manual installation, see the capture-skills repository.

Usage

Once installed, Claude will automatically use the Capture skill when you ask it to:

  • Take a screenshot of a website
  • Generate a PDF from a URL
  • Extract content from a page
  • Get metadata from a URL
  • Create an animated recording of a page

Example prompts:

"Take a screenshot of https://example.com"
"Create a full-page screenshot of news.ycombinator.com in dark mode"
"Screenshot this page as it would appear on an iPhone"
"Generate an A4 PDF of this article with backgrounds"
"Create a landscape PDF of this page"
"Extract the content from this blog post as markdown"
"Get the metadata from https://example.com"
"Record a 10 second GIF of this page"

Available Commands

The skill provides access to all Capture CLI commands:

CommandDescription
capture screenshot <url> -o <file.png>Capture screenshot
capture pdf <url> -o <file.pdf>Generate PDF
capture content <url> --format markdownExtract content
capture metadata <url> --prettyGet metadata
capture animated <url> -o <file.gif>Record animated GIF

For the full list of options, see the capture-skills repository or the individual option pages:

On this page