Search Results for "human"

PentestGPT is an AI-powered autonomous penetration testing agent designed to perform intelligent, end-to-end security assessments using large language models. Published at USENIX Security 2024, it combines advanced reasoning with an agentic workflow to automate tasks traditionally handled by human pentesters. The platform supports multiple penetration testing categories, including web security, cryptography, reversing, forensics, privilege escalation, and binary exploitation. PentestGPT runs in a Docker-first environment, providing a secure, reproducible setup with built-in tooling and session persistence. It offers real-time feedback and live walkthroughs, allowing users to observe each step of the testing process as it unfolds. ...

...It combines artificial intelligence with traditional penetration testing tools to create a fully autonomous pipeline capable of discovering vulnerabilities and executing security assessments without human intervention. It begins with a multi-phase reconnaissance engine that maps the entire attack surface of a target, collecting information such as subdomains, open ports, services, and potential vulnerabilities. RedAmon then uses an AI agent orchestrator to analyze this data, select appropriate tools, and perform exploitation steps such as credential brute forcing or CVE-based attacks. ...

This project explores USB device emulation attacks—commonly called BadUSB—by demonstrating how commodity USB hardware can impersonate keyboards, network adapters, or storage devices to perform scripted actions on a host. It typically contains firmware examples, payloads, and explanations showing how a device presenting as a Human Interface Device (HID) can inject keystrokes, open shells, or orchestrate data exfiltration when plugged into a machine. The codebase is frequently intended for security research and defensive testing: defenders and red teams use it to validate endpoint controls, USB whitelisting, and user training. Due to the dual-use nature of such techniques, responsible repositories emphasize lab-only experiments, consent-based testing, and mitigations like disabling autorun, enforcing device policies, and using endpoint detection.

GitGot is an open source security tool designed to help users quickly search large amounts of public data on GitHub to identify potentially exposed secrets. It operates as a semi-automated, feedback-driven system that combines automated search capabilities with human guidance to refine results during investigation. GitGot leverages the GitHub Search API to perform queries across repositories, files, and gists, allowing security researchers and penetration testers to discover sensitive information that may have been unintentionally exposed in public code. During a search session, users review results and provide feedback that allows GitGot to filter out irrelevant or repetitive findings. ...

...Because the behaviors are stable and documented, teams can run comparative tests over time and quantify regression or improvement in their pipelines. It’s equally useful for human training, giving analysts a safe playground to practice exploitation and triage skills.

SourceAFIS is a software library for human fingerprint recognition. It can compare two fingerprints 1:1 or search a large database for matching fingerprint. It takes raw fingerprint images on input and produces matching score on output. Click the website link above for more info.

...The project uses the fiwalk.py package to extract relevant artifacts from a disk image in the DFXML format, performs an analysis and saves the results in a form for human readers.

Most often we input wrong passwords due to CAPS Lock or Typo. The solution is to reduce the server communication for such mistakes. The purpose of the project is to capture and address unintentional human errors that could be resolved at the client side during the login process. Solution Abstract: An array of images are loaded at the client side. One of the images is displayed for every password entered based on a simple algorithm. The image is computed based on the password and hence it will change only when the user changes his password. ...

This class is intended to block the action of spam bots or robots (as Captcha), employing inteligent tactics on human detection. Made on PHP.

Vernash: Stream cipher based on the Vernam cipher and Variable-Length Hashes

GigoIt’s HumanAuth supports ADA and Section 508 requirements, increased security and includes watermarked images with random positioning. HumanAuth ensures that an actual human is using your site without forcing them to read distorted CAPTCHA text.

TURing human IDentification, a textual highly accessible alternative to image CAPTCHAs involving the usage of simple phrases and based on the language recognition features of the user, supposedly human.

HumanSudokuSolver is intended to solve Sudoku puzzles in a way human beings would do (non brute force). It currently solves most of the puzzles I tried and can output a step-by-step solution. Developers who want to contribute are welcome!

The Nemesis Project is designed to be a command line based, portable human IP stack for UNIX-like and Windows systems. The suite is broken down by protocol, and should allow for useful scripting of injected packets from simple shell scripts.

The development of a biometric system and applications e.g. Access Control based on the system to recognize and verify people. The primary goal is face recognition, but other human attributes might get used too in the future.