skills.rest
salomepoulain
salomepoulainCOMMUNITY
Explore all skills by this author →
GitHub

browse

Automate real browser interactions through a camofox-browser HTTP server with accessibility-tree snapshots.

Updated Sep 8, 2026
Search Tags:#browser automation#web scraping#firefox#pdf download#accessibility tree#youtube transcript#camofox
One-click install
npx skills add https://github.com/salomepoulain/makery-stations --skill browse-salomepoulain
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browse
Source: https://github.com/salomepoulain/makery-stations/tree/main/stations/claude/workbench/pantry/skills/browse
Command: npx skills add https://github.com/salomepoulain/makery-stations --skill browse-salomepoulain

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires camofox-browser, camoufox-js, yt-dlp.

What problem does it solve? Static fetching tools like WebFetch fail on JavaScript-rendered pages, login flows, and sites that serve PDFs through JS viewers. This Skill drives a real anti-detection Firefox instance via the camofox-browser HTTP server, letting you read pages, click elements, type text, and download files through a simple REST API. ## Core Features & Use Cases - Interactive browsing: Create tabs, navigate, take accessibility-tree snapshots with stable element refs, then click, type, scroll, and press keys against those refs. - Academic paper retrieval: Navigate publisher sites (ACM, arXiv, journal pages), dismiss modals, and capture PDF downloads server-side even when the file sits behind a JS viewer. - Search macros and transcripts: Use built-in macros like @google_search and @reddit_search, and fetch YouTube transcripts via a dedicated endpoint. - Use Case: You need a paper whose PDF only appears inside ACM's eReader. Open the DOI page, click the page-level download link, and pull the file bytes from the server's downloads endpoint as base64. ## Quick Start Ask the agent to start the camofox-browser server and search Google for a topic, then read and summarize the first result page.

Frequently Asked Questions about browse

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I browse JavaScript-rendered pages with an AI agent?▼

Use the camofox-browser HTTP server, which drives a real Firefox instance and returns accessibility-tree snapshots with stable element refs. Create a tab, navigate to the URL, take a snapshot, then click or type using the returned refs.

How to download academic papers from ACM or arXiv programmatically?▼

Navigate to the DOI or article page, dismiss any modal, and click the page-level download link rather than the PDF viewer's save button. The server captures the download, and you fetch the bytes as base64 from the tab's downloads endpoint.

Why does navigating to a PDF URL fail with 'Download is starting'?▼

This error is expected behavior, not a real failure. The server has already captured the file server-side, so list the tab's downloads endpoint and fetch the content with includeData=true to retrieve the base64 bytes.

Can I use this for sites that require login sessions?▼

Yes, sessions persist per userId with cookies and storage isolated between callers. You can also import Playwright-format cookies via the sessions endpoint and export or reset stored session state.

When should I use a real browser instead of WebFetch?▼

Use a real browser when pages require JavaScript rendering, logins, or multi-step click flows, since static fetching only retrieves raw HTML. For plain static pages, a simple fetch is faster and sufficient.

Related Skills

Explore More Software Engineering Skills→
alexchloualexchlou
2

research-start

Generate structured legal research plans with statutes and search queries.

Community
Basic
chiragborse1chiragborse1

summarize

Summarize web pages, PDFs, videos, and local documents into concise highlights.

Community
Intermediate
izzoaizzoa
22

ss-verify

Inspect rendered browser screenshots for visual defects in layout, typography, contrast, and spacing.

Community
Advanced
marktantongcomarktantongco

fetch

Fetch HTML or JSON from static web pages with redirect and proxy control.

Community
Basic