#1212th of 14 in Security
pwnagotchi
WiFi handshake-capturing agent that tunes its attacks with reinforcement learning
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 9.2k
- License
- custom license
- Last commit
- Aug 2026
- Last release
- Apr 2021
- Language
- Python
8,948 on 10 Apr 20269,246 on 30 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Pwnagotchi runs an A2C reinforcement-learning agent (LSTM policy with MLP feature extractor) on top of bettercap to capture WPA key material from nearby WiFi. It saves PMKIDs and full or half handshakes as PCAP files that hashcat can crack. It adjusts its parameters over epochs, and nearby units can split channels between them.
Who it is for: Security hobbyists auditing WPA WiFi networks they own or are authorized to test
Strengths
- Captures PMKIDs and full or half WPA handshakes as PCAPs usable by hashcat
- Agent tunes its own parameters per environment over time using A2C with an LSTM policy
- Multiple units detect each other and divide channels between them
- Docs, forum and community channels are listed on the project site
Weaknesses
- Last tagged release is 2021-04-18, though commits continue
- License detected as NOASSERTION, while the README states GPL3
- Performs poorly at first and needs time in varied WiFi environments to learn
- README gives no hardware requirements, RAM figures or supported boards
What it needs
- Needs bettercap, hashcat
- Models: A2C (stable-baselines, MlpLstmPolicy)
Quick answers
Is pwnagotchi open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Also in Security
See all 14| Rank | Project | Score |
|---|---|---|
| 1 | shannonAutonomous pentester that exploits web apps and APIs using source code analysis | 91 out of 100 |
| 2 | ifixaiPython CLI that audits AI agents for policy, access and reliability failures | 86 out of 100 |
| 3 | StrixAutonomous AI pentesting agents that validate vulnerabilities with working exploits | 84 out of 100 |
| #4 | pentagiAutonomous multi-agent penetration testing system with sandboxed security tools | 78 out of 100 |
| #5 | ai-infra-guardRed-teaming platform for AI infra CVEs, MCP servers, agent skills and jailbreaks | 71 out of 100 |