Camofox Browser

Active
GitHub JavaScript MIT

Description

Camofox Browser is a headless browser automation server powered by Camoufox, a Firefox fork with C++-level fingerprint spoofing. It bypasses Google, Cloudflare, and most bot detection, providing token-efficient accessibility snapshots and stable element references for AI agents.

Key Features

  • C++ level fingerprint spoofing bypassing bot detection
  • Stable element references (e1, e2, e3) for reliable interaction
  • Token-efficient accessibility snapshots (~90% smaller than HTML)
  • Search macros for Google, YouTube, Amazon, and Reddit
  • Session isolation with separate cookies and storage
  • Proxy and GeoIP support with automatic locale/timezone

Use Cases

💡 AI agent web browsing without detection
💡 Web scraping with anti-bot protection bypass
💡 Automated testing on sites with fingerprint protection
💡 Content extraction from protected websites
💡 Social media monitoring and data collection

Strengths & Limitations

Strengths

  • Actively maintained, recent updates
  • High community interest (10.5k stars)
  • Permissive open-source license (MIT)

Quick Start

Run npx @askjo/camofox-browser or clone and npm install && npm start. Server starts at http://localhost:9377. Use REST API with accessibility snapshots for agent interaction. Supports Docker deployment via Makefile.

Related Projects

Browserable

1.2k · JavaScript
Stale C

Browserable is a self-hostable browser automation tool purpose-built for AI agents. It provides secure Docker-based browser environments with a JavaScript SDK, achieving 90.4% accuracy on the Web Voyager benchmark for autonomous web navigation.

browser-automationself-hosteddocker +3
  • · Open-source browser automation library purpose-built for AI agents, supporting site navigation, form filling, and data extraction
  • · Achieves 90.4% accuracy on the Web Voyager benchmark, leading in performance
  • · Docker-based self-hosted architecture providing secure isolated browser environments

Moli

1.9k · Rust
Active A

A production-ready headless browser for AI agents that treats the DOM as the source of truth and triggers layout or paint only on demand, usable via CLI, CDP, and WebDriver protocols.

headless-browserbrowser-automationweb-scraping +3
  • · Standalone browser kernel — a Rust engine with its own V8, Stylo, and Taffy layout integration, not a Chromium wrapper
  • · On-demand rendering — skips layout and paint by default so structure-first requests stay fast and light on memory
  • · Extraction-friendly output — the CLI emits HTML, Markdown, JSON, and semantic trees with selector and network waits