Traktor Web Scraper
PassAudited by ClawScan on May 10, 2026.
Overview
This appears to be a disclosed browser-based website scraper, with the main cautions being parallel background browser automation and saving content from any URLs the user provides.
Install only if you are comfortable using a browser-extension-based scraper. Run it on URLs you are allowed to scrape, keep URL batches small, and use a separate or logged-out browser profile if you do not want authenticated or private page content saved into your project.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Running it on many URLs may open multiple browser tabs/tasks and create a large number of web requests and local files.
The skill intentionally launches a background extraction agent for every supplied URL; this is aligned with its parallel scraping purpose, but it can consume resources or generate many site requests if used on many URLs.
For EACH URL, call the Task tool ... run_in_background: true
Use a small, intentional URL list, monitor background tasks, and only scrape sites where you have permission.
If the user supplies a private, logged-in, or internal URL, the skill may save page content and assets from that browser session into the project folder.
The skill operates through an MCP browser extension and browser context. This is disclosed and purpose-aligned, but it means extraction happens through the user's browser environment.
Requires Claude Code with claude-in-chrome MCP server (browser extension).
Use a separate browser profile or logged-out session for public scraping, and avoid private or sensitive pages unless you intend to save their content locally.
The agent will execute DOM-inspection JavaScript in the opened web pages to collect images, videos, SVGs, fonts, metadata, and page structure.
The skill runs fixed JavaScript snippets in the browser page to inspect the DOM and discover assets. This is expected for a browser scraper, but users should understand that page-context scripting is part of the workflow.
Call mcp__claude-in-chrome__javascript_tool with this code:
Use it on pages you are comfortable inspecting through automation, and review the extraction behavior if handling sensitive sites.