Javbus

Javbus Scraper

Overview

This skill enables accessing javbus.com, navigating through age verification challenges using AI, and extracting video links from the site.

Workflow

Step 1: Open Website

Navigate to the target URL using agent-browser:

agent-browser open https://javbus.com

Step 2: Check for Age Verification

Take a snapshot to see the current page state:

agent-browser snapshot

Look for age verification elements such as:

• "I am 18+" buttons
• Age entry forms
• "Enter" or "Confirm" buttons
• Warning text about adult content

Step 3: Handle Age Verification

If age verification is present:

1. Analyze the page content from the snapshot
2. For age entry questions, use AI to determine the correct response:

• If asked "Are you 18 or older?", click the "Yes" option
• If asked to enter age, input 18 or above
• Look for buttons like "Enter", "I Agree", "Confirm"

Use appropriate agent-browser commands:

agent-browser click      # For buttons
agent-browser fill  "18" # For age entry fields
agent-browser press Enter          # Submit the form

1. Take another snapshot to verify success:

agent-browser snapshot

If no age verification is present:

Skip to Step 4.

Step 4: Extract First Video Link

After successfully passing age verification (or if no verification was needed):

1. Take a snapshot of the main page:

agent-browser snapshot

1. Look for video link elements, typically with:

• `` tags pointing to video pages
• Video thumbnails or titles
• Classes or attributes containing "movie", "video", "item"

1. Extract the first video link using:

agent-browser get attr href 

1. Print the first video link to verify successful navigation

Tips

• Use agent-browser snapshot -i for a compact view of interactive elements only
• The @ref system in snapshots allows you to reference elements (e.g., @e2 refers to element #2)
• Use agent-browser get text to read page content if needed for analysis
• If age verification fails, take a snapshot and analyze what went wrong

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: buhe
• Source: buhe/claude-common-skills
• License: MIT

Install and usage instructions live in the source repository linked above.