evilsocket/pwnagotchi

(⌐■_■) - Deep Reinforcement Learning instrumenting bettercap for WiFi pwning.

What it solves

Pwnagotchi is designed to automate the collection of crackable WPA key material (such as PMKIDs and full or half WPA handshakes) from surrounding WiFi own-networks. It removes the need for manual tuning of attack parameters to optimize the capture of these handshakes in different environments.

How it works

The project uses an Advantage Actor-Critic (A2C) reinforcement learning agent with an LSTM and MLP feature extractor as its policy network. By leveraging bettercap, the agent learns from its WiFi environment over time (measured in epochs) by tuning its own parameters to maximize the amount of captured key material. It can also communicate with other Pwnagotchi units via a custom parasite protocol on top of the dot11 standard, allowing multiple units to divide channels among themselves to cooperate for optimal capture efficiency.

Who it’s for

It is intended for security researchers and WiFi own-network penetration testers who want an autonomous, learning-based tool for capturing WPA handshakes.

Highlights

  • A2C-based reinforcement learning for autonomous parameter tuning.
  • Captures PMKIDs and full/half WPA handshakes as PCAP files.
  • Cooperative behavior: multiple units can coordinate channel division among themselves.
  • Uses an LSTM with MLP feature extractor for its policy network.

Related

  • Dispatch
    Advantage Actor Critic (A2C) Explained

    Hugging Face explains Advantage Actor Critic (A2C), a hybrid reinforcement learning architecture that reduces variance in policy gradient estimation by combining policy-based and value-based methods.

  • Project
    Farama-Foundation/Gymnasium-Robotics

    A collection of Reinforcement Learning robotic environments powered by the MuJoCo physics engine and the Gymnasium API for training AI agents in manipulation and navigation tasks.

  • Dispatch
    OpenAI RL²: Fast Reinforcement Learning via Slow Reinforcement Learning

    OpenAI introduces RL², a method that uses a recurrent neural network (RNN) to learn a reinforcement learning algorithm, allowing agents to adapt to new tasks in a few trials by encoding the learning process itself into the RNN weights.

  • Project
    greyhaven-ai/autocontext

    A recursive self-improvement harness that helps AI agents iteratively improve their performance on tasks by running evaluations and distilling lessons into playbooks and datasets.

  • Project
    vxcontrol/pentagi

    An autonomous AI-powered penetration testing platform that uses a multi-agent system to automate security vulnerability research and exploitation in isolated environments.