camofox-browser

Use it when

• Your agent's Playwright or headless Chrome sessions get blocked by Cloudflare, Google, or fingerprint-based bot detection despite stealth plugins
• You need a shared, low-memory browser backend for AI agents with per-user session isolation, cookie import for authenticated sites, and token-efficient accessibility snapshots instead of raw HTML

What it solves

Not the fit when

• Circumventing site terms, authentication, or access controls you are not authorized to bypass
• High-volume distributed crawling; the documented architecture uses one browser instance with per-user concurrency controls
• Chromium-specific capabilities such as session video recording
• Assuming anti-detection is guaranteed against every site; bypass and resource figures are project claims, not universal guarantees
• Deployments where anonymous crash/hang telemetry is unacceptable unless CAMOFOX_CRASH_REPORT_ENABLED=false is set
• structured extraction from static HTML
• PDF or document parsing
• Chromium-specific features such as recordVideo
• native mobile app automation

Install

npx @askjo/camofox-browser (or git clone https://github.com/jo-inc/camofox-browser && cd camofox-browser && npm install && npm start; Docker via make up)

Invoke

curl -X POST http://localhost:9377/tabs -H 'Content-Type: application/json' -d '{"userId":"agent1","sessionKey":"task1","url":"https://example.com"}'; then GET /tabs/:id/snapshot and POST /tabs/:id/click with element refs

Alternatives

No reviewed alternatives recorded yet.