Skip to content
Kai Piper edited this page Feb 20, 2026 · 5 revisions

DuckHunter

Prevent RubberDucky (or other keystroke injection) attacks

Try out the new setup GUI, which helps you to set up the software, and we have just released a new feature that allows you to run the script every time your computer starts automatically

Read this program's postmortem at my blog

Intro

[Rubberduckies](https://hakshop.myshopify.com/products/usb-rubber-ducky-deluxe) are small USB devices that pretend to be USB keyboards and can type on their own at very high speeds. Because most -if not all- OS trust keyboards automatically, it is hard to protect oneself from these attacks.

DuckHunt is a small, efficient script that acts as a daemon consistently monitoring your keyboard usage (right now, speed and selected window) that can catch and prevent a rubber ducky attack. (Technically, it helps prevent any type of automated keystroke injection attack, so things like Mousejack injections are also covered.)

Features

Protection Policy

  • Paranoid: When an attack is detected, keyboard input is disallowed until a password is entered. Attack will also be logged.
  • Normal: When an attack is detected, keyboard input will temporarily be disallowed. (After it is deemed that the treat is over, keyboard input will be allowed again. Attack will also be logged.
  • Sneaky: When an attack is detected, a few keys will be dropped (enough to break any attack, make it look as if the attacker messed up.) Attack will also be logged.
  • LogOnly: When an attack is detected, simply log the attack and in no way stop it.

Extras

  • Program Blacklist: If there are specific programs you never use (cmd, PowerShell). Consider interactions with them as highly suspicious and take action based on the protection policy.
  • Support for AutoType software (eg. KeePass, LastPass, Breevy)
  • Whitelist support for trusted windows/workflows that should bypass checks.
  • Advanced burst detection (rapid interval streak and injected-event streak).
  • Optional signature-pattern detection for suspicious command-launch sequences.
  • Optional adaptive threshold mode (learned baseline + blended threshold).
  • Optional low-variance burst detection for machine-like typing cadence.
  • Optional per-window threshold overrides for finer tuning by application.
  • Temporary lockout timer in Normal mode to better absorb attack bursts.
  • Structured intrusion logs with reason + context for easier analysis.
  • Live runtime status telemetry (optional JSON export) and pause/resume controls.
  • Rotating log support to cap disk usage on long-running installs.
  • Optional warmup calibration mode to reduce startup false positives.

Setup

Regular users:

Advanced Users

  • Keep Reading...
  • Feel Free to contact me, add issues, fork, and get involved with this project :). Together we can make a stronger tool!

Requirements

Advanced Setup

  • Step 1. Customize duckhunt.conf variables to your desire
    • You can customize the password, speed threshold, privacy, etc.
    • You can also tune advanced protection variables: normal_lockout_ms, rapid_burst_interval_ms, rapid_burst_count, injected_burst_count, whitelist, pattern_signatures, adaptive_threshold_enabled, adaptive_*, low_variance_*, window_threshold_overrides, status_filename, log_max_bytes, and warmup_*
  • Step 2. Turn the duckhunt-configurable**.py** to a duckhunt-configurable**.pyw** so that the console doesn't show up when you run the program
  • Step 3. (opt) Use Py2Exe to create an executable.
  • Step 4. Run the program. You are now protected from RubberDuckies!

TODO

  • More monitoring features:
  • Add OSX & Linux support!
  • Look for certain patterns (eg, "GUI D, GUI R, cmd, ENTER")
  • Quality of life updates

Happy Hunting!

Imagine you are at your desk working in your computer, you stand up to pick a cup of coffee from the nearby coffee machine. In that time, a sneaky someone gets up plugs a strange pendrive into your computer, waits ten seconds, and walks away. They just compromised your computer in a blink of an eye a la Mr. Robot style.

Clone this wiki locally