nowhere.cool

Back in 2020, I stumbled across an interesting device called Pwnagotchi while looking to christen my first raspberry pi zero I picked up. The open source project piqued my interest as it had the right mix of complexity and I felt ready to dive in the wide wide world of edge computing, IoT networks and ARM/GPIO. In hindsight, the project really opened my eyes to the amount of networking across the device connected ecosystem and propelled me further into computer electronics, WiFi and deeper into more of the radio spectrum with technology like HAM, LORA(WAN), GPRS, 4/5G and HaLOW. Fast forward 6 years later, here are all the different build variations I’ve experimented with.

What is it?

A tiny, pocket-sized device that “listens” to nearby Wi-Fi networks and learns how to capture data from them passively. It doesn’t connect to anything—it just watches the radio waves in the air. Think of it like a weather station, but instead of measuring wind and temperature, it measures Wi-Fi signals around you.

What It Does

• Listens for Wi-Fi devices nearby – Phones, laptops, smart TVs… anything speaking Wi-Fi.
• Collects handshake data – This is the “hello” exchange between a device and a Wi-Fi network. It’s not the password—it’s just data that can be analyzed later.
• Learns from its environment – It uses a simple machine-learning loop to improve the way it captures these signals.
• Shows a Tamagotchi-like face – It has a digital pet personality that reacts to how well it’s doing.

What It’s For

Primarily, education – it’s a fun way to learn about Wi-Fi, wireless security, and penetration testing. Also, Security research to better understand how networks behave and where their weaknesses might be.

---

Not all Pwnagotchi are alike

The community and device continues to grow with tons of plugins as well as a few different flavors of the pwnagotchi

Forks

• (OG) evilsocket – https://pwnagotchi.ai
• aluminum-ice
• jayfelony – https://pwnagotchi.org/

Community

• https://www.reddit.com/r/pwnagotchi/
• 3rd Party/Community Plugins

---

My Builds

Build 1:

My first build is based off of the original version from evilsocket. I kept it pretty simple with only 2 plugins and opted for a pre-built case at the time as I didn’t have a printer back then. The most notable feature of the OG version is the use of A2C reinforcement learning for efficient tuning of the devices methods. There is some debate of its stability (and its RPi0W 32bit only) but still an incredibly clever implementation of AI for its time.

---

Build 2, Variation 1:

The second is built off of jayofelony’s version. While I’m currently satisfied at the performance of it in its current state, it’s still quite bulky, runs pretty hot when charging and the display is already wearing at the edge where it connects to GPIO.

---

Build 2, Variation 2:

This variation swaps the GeekWorm UPS for a Waveshare hat with a 1000mah battery. The look and feel was pretty much there for a compact, discrete build but the tradeoff here is battery life for overall size.

---

Build 3, Variation 1

Hardware

• Raspberry Pi 5 64bit
• Adafruit MiniPiTFT Screen
• UPS Hat C for Pi
• 32GB SD Card
• Alfa AWUS036ACS USB WiFi dongle

Software

• jayofelony image (64bit)
  • modded Pwnagodchi config