camofox-stack

Automates stealth web scraping and browser interaction using the CamoFox anti-detection browser stack.

Updated Sep 2, 2026
One-click install
npx skills add https://github.com/nbiish/tts-mcp --skill camofox-stack-nbiish
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: camofox-stack
Source: https://github.com/nbiish/tts-mcp/tree/main/.agents/skills/camofox-stack
Command: npx skills add https://github.com/nbiish/tts-mcp --skill camofox-stack-nbiish

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Standard browser automation gets blocked by bot detection systems like Cloudflare and Datadome through TLS and canvas fingerprinting. This Skill guides agents through the three-layer CamoFox stack (camoufox engine, camofox-browser server, camofox-mcp bridge) to perform web scraping and automation that evades fingerprint-based detection. ## Core Features & Use Cases - Three-Layer Architecture Guidance: Explains how the camoufox C++ Firefox fork, the camofox-browser REST/CLI server, and the camofox-mcp agent bridge interlock for stealth automation. - CLI Automation Patterns: Provides shell-ready commands for opening tabs, capturing tabId values, taking accessibility snapshots, and interacting via [eN] element refs. - Structured Extraction & Auth Vault: Uses schema-based extractStructured calls instead of injected JavaScript, and injects credentials via the camofox auth vault without exposing passwords to the LLM. - Use Case: Scrape a Cloudflare-protected site by starting camofox-browser, opening the URL with the CLI, snapshotting the accessibility tree, clicking elements by ref, and extracting structured JSON data. ## Quick Start Use the camofox skill to open a Cloudflare-protected URL, snapshot the page, and extract the article title as structured JSON.

Frequently Asked Questions about camofox-stack

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

FAQPage Schema
How do I bypass Cloudflare bot detection for web scraping?▼

Use the CamoFox stack, which runs a C++ Firefox fork that spoofs TLS and canvas fingerprints at the engine level to defeat Cloudflare and Datadome. Start camofox-browser, open the target URL via the CLI, snapshot the accessibility tree, and interact using element refs.

How to use the camofox CLI for browser automation?▼

Run camofox open with a URL and capture the returned tabId, then pass that tabId to snapshot, click, and type commands along with a userId. Always capture the tabId explicitly, since relying on implicit active tab state causes 'No active tab found' errors in fast scripts.

Can CamoFox bypass IP bans on Google Scholar or JSTOR?▼

No, CamoFox only defeats fingerprint-based bans, not IP reputation bans. For sites blocking datacenter IPs, route traffic through residential proxies using CAMOFOX_PROXY_PROFILES_FILE or the --proxy-host argument, since no browser engine can spoof an IP address.

How many browser sessions can CamoFox run on 16GB RAM?▼

Limit concurrent isolated userId contexts to 5-8 on a 16GB machine, since each Camoufox context uses about 1.2GB of RAM. Additional tabs within the same userId context cost only 50-100MB each, so reuse contexts when you need more tabs.

Why does the camofox REST API /tabs endpoint fail?▼

The /tabs endpoint requires both a userId and a sessionKey in the JSON body, and missing either causes failures. The sessionKey controls session reuse scope and proxy bindings, so include both fields plus the url in every POST request.