ush.py is an fork of HTTPshell.py. It is a single-file, dependency-minimal remote shell that operates entirely over Websocket. It is specifically optimized for extremely resource-constrained environments (like 32MB RAM) and networks where inbound TCP access is restricted by GCNAT or aggressive firewalls.
By leveraging a PTY-based approach and smart backpressure management, ush.py provides a full interactive terminal experience over standard web traffic.
Important
ush.py IS EXTREMELY UNSECURE. Traffic is sent in plain text. If there is a sniffer or a Man-In-The-Middle (MITM) on the network, your credentials and session data will be stolen.
If you require built-in encryption or even more security, use HTTPshell.py instead. ush.py should only be used over trusted local networks or behind a secure HTTPS reverse proxy (like Caddy or Cloudflare).
ush.py comes with ABSOLUTELY NO WARRANTY, to the extent permitted by applicable law.
Many networks (ie: GCNAT) block inbound connections. This makes SSH or other remote shell protocols unusable. And cloudflare has it on a paywall.
ush.py solves this by multiplexing a full TTY session over standard Websocket, making it compatible with almost any reverse proxy (Cloudflare, Caddy, Nginx) and bypassing inbound port restrictions.
- Python 3.8+
The server side only supports Linux. The client side supports absolutely anything.
Read the releases, there will be a install guide in there.
Run the server on the remote machine. It must be run as root to access /bin/login for PAM authentication.
sudo ush --server -p 8080On your local machine, connect using the server's host/IP.
ush <host> -p 8080*To exit the session, use the shortcut: Ctrl + ]*
- PTY Powered: Supports
vim,htop,top, and full shell interactivity with proper terminal resizing. - Fast: Highly optimized code for minimal overhead.
- Proxy Friendly: Designed to work flawlessly behind Cloudflare, Caddy, and other reverse proxies.
ush.py aims to be the smallest, most reliable HTTP shell possible. Pull requests regarding further code compression, or better memory management are highly encouraged.