{"id":39088,"date":"2019-10-31T22:27:47","date_gmt":"2019-10-31T19:27:47","guid":{"rendered":"https:\/\/prohoster.info\/blog\/pervyj-vypusk-pwnagotchi-igrushki-dlya-vzloma-wifi-setej\/"},"modified":"2019-10-31T22:27:47","modified_gmt":"2019-10-31T19:27:47","slug":"pervyj-vypusk-pwnagotchi-igrushki-dlya-vzloma-wifi-setej","status":"publish","type":"post","link":"https:\/\/prohoster.info\/en\/blog\/news\/pervyj-vypusk-pwnagotchi-igrushki-dlya-vzloma-wifi-setej","title":{"rendered":"The first release of Pwnagotchi, a toy for hacking WiFi networks","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Introduced<\/a><\/noindex> The first stable release of the project Pwnagotchi<\/a><\/noindex>, developing a tool for hacking wireless networks, designed as an electronic pet reminiscent of a Tamagotchi toy. The main prototype of the device is built<\/a><\/noindex> is built on a Raspberry Pi Zero W board (provided firmware<\/a><\/noindex> for booting from an SD card), but it can also be used on other Raspberry Pi boards or any Linux environment with a wireless adapter that supports monitoring mode. Control is performed via an attached LCD screen or through web interface<\/a><\/noindex>. The project code is written in Python and is distributed<\/a><\/noindex> under the GPLv3 license.<\/p>\n

To keep the pet in good spirits, it needs to be fed with network packets sent by participants of wireless networks during the handshake phase of a new connection. The device finds available wireless networks and attempts to intercept handshake sequences. Since the handshake is sent only when a client connects to the network, the device uses various techniques to drop current connections and force users to reconnect to the network. Throughout the interception, a database of packets accumulates, containing hashes that can be used for WPA key cracking.
\n

\"The<\/a><\/noindex><\/center><\/p>\n

The project is notable for using methods of reinforcement learning<\/a><\/noindex> AAC (Actor Advantage Critic) and a neural network based on long short-term memory<\/a><\/noindex> (LSTM) that have gained popularity in creating bots for progressing through computer games. The training of the learning model occurs during the device's operation, taking into account past experiences to select the optimal strategy for attacking wireless networks. Through machine learning, Pwnagotchi dynamically adjusts traffic interception parameters and chooses the intensity of forcibly terminating user sessions. A manual mode of operation is also supported, wherein the attack is executed 'head-on'.<\/p>\n

To intercept the types of traffic necessary for WPA key cracking, the bettercap<\/a><\/noindex>package is used. Interception occurs both in passive mode and through known types of attacks that force clients to resend identifiers PMKID<\/a><\/noindex>. The intercepted packets cover all forms of handshake supported in hashcat<\/a><\/noindex>, stored in PCAP files calculated, one file for each wireless network. <\/p>\n

\"The<\/a><\/noindex><\/center><\/p>\n

Similar to Tamagotchi, it supports the identification of nearby devices, and optionally, participation in creating a shared coverage map. The connection of Pwnagotchi devices via WiFi uses the protocol Dot11<\/a><\/noindex>. Nearby devices exchange data about wireless networks and organize joint operations by sharing channels for conducting attacks.<\/p>\n

The functionality of Pwnagotchi can be expanded through plugins<\/a><\/noindex>, which implement features such as automatic software update system, backup creation, linking captured handshakes to GPS coordinates, publishing data about cracked networks on services like onlinehashcrack.com, wpa-sec.stanev.org, wigle.net and PwnGRID<\/a><\/noindex>, additional indicators (memory consumption, temperature, etc.) and performing dictionary attacks for intercepted handshakes.<\/p>\n