| name | camoufox |
| description | Use for browser automation through camoufox-browser: open a page, inspect it with snapshot refs, click, fill, select, upload, take screenshots, and drive websites through the CLI instead of direct browser code.
|
Camoufox Browser Automation
Use camoufox-browser for browser automation. Prefer the CLI directly. The MCP
server is just a thin adapter over the same command layer, and this skill exists
to teach the agent the normal CLI workflow.
Primary workflow
- Open a page:
camoufox-browser open https://example.com
- Inspect the page and collect refs:
camoufox-browser snapshot
- Interact using refs from the snapshot:
camoufox-browser click 'button:Sign in'
camoufox-browser fill 'textbox:Email' user@example.com
camoufox-browser select 'combobox:Country' Ukraine
camoufox-browser upload 'textbox:Resume' /absolute/path/resume.pdf
-
Re-run snapshot after navigation or significant DOM changes.
-
Capture artifacts when needed:
camoufox-browser screenshot --output page.png
camoufox-browser eval '() => document.title'
Notes
- You do not need to run
start for the normal flow. camoufox-browser auto-starts the daemon with default settings when a browser command needs it.
- Use
camoufox-browser start --headless only when you need explicit daemon control or startup options like --proxy.
camoufox-browser is intended for Linux and macOS hosts.- Use
camoufox-browser --help to discover the full command surface.
- Install this skill through
npx skills add <repo> --skill camoufox.
Core commands
camoufox-browser open <url>
camoufox-browser snapshot
camoufox-browser click <ref>
camoufox-browser fill <ref> <text>
camoufox-browser select <ref> <value> [more-values...]
camoufox-browser upload <ref> <file> [more-files...]
camoufox-browser screenshot [--output FILE]
camoufox-browser close
Advanced commands
camoufox-browser start [--headless] [--os windows|macos|linux] [--proxy PROXY]
camoufox-browser stop
camoufox-browser status
camoufox-browser tabs list|new|close [index]|select <index>
