~/hackyfeed $

BlueSpy is a proof-of-concept tool designed to record and replay audio from Bluetooth devices without the users' awareness, exploiting vulnerabilities in Bluetooth security protocols. It aims to highlight the insecure use of Bluetooth technology and the necessity for comprehensive security assessments, as outlined in the Bluetooth Security Assessment Methodology (BSAM). Key features include the capability to pair and connect to audio-capable Bluetooth devices, along with utilizing various Linux tools such as `bluetoothctl`, `btmgmt`, and PulseAudio utilities for audio management.

The MCP Server for Binary Ninja facilitates advanced binary analysis by exposing various functions and resources through a Model Context Protocol (MCP) interface. It primarily allows users to interact with Binary Ninja's capabilities, such as renaming symbols and generating pseudocode, via a configurable server running either in the UI or headless mode. Notable features include support for multiple transport methods for clients, a variety of analytical tools, and extensive resource access to enhance binary analysis workflows.

The Bible Databases repository offers a comprehensive collection of Bible versions in various formats including MySQL, SQLite, CSV, JSON, YAML, and more, facilitating easy access and integration for developers and researchers. Its primary use case is to provide a structured database of Scripture texts along with a robust cross-reference system sourced from OpenBible.info, allowing for detailed study and analysis of biblical texts. Notable features include a wide array of translations, conversion scripts, and documentation for seamless implementation and utilization.

A utility for arming (creating) many bees (micro EC2 instances) to attack (load test) targets (web applications).

Basic Pitch is a Python library designed for Automatic Music Transcription (AMT), leveraging a lightweight neural network developed by Spotify's Audio Intelligence Lab. The tool efficiently generates MIDI files from audio inputs, supporting multipitch detection and offering instrument-agnostic transcription capabilities, making it suitable for polyphonic instruments. Notable features include its ease of installation, compatibility with various operating systems, and the ability to handle richer audio inputs while maintaining pitch accuracy.

The "Awesome LLM Apps" repository provides a collection of over 100 open-source AI agents, skills, and Retrieval-Augmented Generation (RAG) applications, all designed to enhance the functionality of various AI models such as Claude, GPT, and Llama. Notable features include easy installation of agent skills via command line, extensive documentation, and a diverse range of pre-built agents capable of tackling tasks from project management to fraud investigation. This tool serves as a practical resource for developers looking to implement AI solutions rapidly and effectively.

The MDC Rules Generator is a community-driven tool designed to create Markdown Cursor (MDC) rule files from structured JSON data containing library information. It utilizes Exa for semantic web search and LLM capabilities (Gemini) for generating detailed content, while supporting parallel processing and a smart retry mechanism for efficiency. Notable features include the ability to process only failed libraries by default, extensive command-line options for customizability, and straightforward integration for adding new libraries.

Awesome Artificial Intelligence is a curated collection of high-quality resources aimed at software developers who seek to learn, build, and deploy generative AI and agentic systems. The repository emphasizes technical depth and practical application, offering valuable materials such as books and courses that cover foundational AI concepts, coding agents, and production workflows. Notably, it features an active curation process that ensures the relevance and quality of included resources.

AutoDroid is a system designed for automating intelligent tasks on smartphones by leveraging large language models (LLMs). Built on the DroidBot framework, it facilitates the execution of app-specific tasks using captured screenshots and view hierarchies from the target applications, enabling users to automate interactions with various Android apps. Notable features include a dataset called DroidTask that provides comprehensive state and task information, along with the ability to specify tasks and outputs through command-line interfaces.

*attrs* is a Python library designed to simplify the creation and management of classes by automating the implementation of object protocols, thereby enhancing both code clarity and correctness. Its primary use case lies in reducing boilerplate code for class definitions by using decorators and declarative attribute declarations, making class creation more concise and efficient. Notable features include automatic generation of special methods (dunder methods) and support for default values and factory functions, streamlining class design for developers.

The ATLAS Navigator Data repository, now deprecated, provided functionality for generating ATLAS STIX data and Navigator layer files for use in the ATLAS Navigator application. Key features included pre-structured Navigator layers for case studies and the ATLAS matrix, STIX 2.1 bundle outputs, and scripts for maintaining updated data. Users are directed to migrate to the newer ATLAS Data repository for continued access to these functionalities.

MITRE ATLAS Data provides a comprehensive repository of adversarial tactics, techniques, and procedures (TTPs) specifically targeting AI systems. The tool facilitates the parsing, validation, and management of ATLAS data, enabling users to access and utilize a structured knowledge base for security assessments of AI technologies. Notable features include a REST API, command-line tools for data integration and distribution in multiple formats, and monthly updates to ensure current relevance.

APIDetector is a sophisticated tool designed for the testing of exposed Swagger endpoints across multiple subdomains, focusing on robust detection of vulnerabilities and minimizing false positives. It features a modern web interface, real-time results, support for bulk domain scanning, and advanced capabilities like automated XSS detection and visual proof of concept generation. The tool is tailored for security professionals and developers aiming to enhance their API security assessment workflows.

angr-management is a cross-platform graphical tool designed for binary analysis, built on the angr binary analysis framework. Its primary use case is to facilitate disassembly and decompilation of binaries, offering an intuitive interface for users to perform complex analyses. Notable features include interactive visualizations of disassembled code and decompiled output, enhancing the usability of binary analysis workflows.

Android Automate is a root-less automation tool designed for Android devices, facilitating multi-touch recording and playback with precise timing. It features an intuitive Python object-oriented API that allows users to inject ADB commands for automating unit and regression testing across both real devices and emulators. Notable functionalities include an easy-to-use CLI for recording touch events, managing emulators, and searching installed applications.

ALPRovingGround is a Python application designed to evaluate adversarial noise attacks on license plate recognition systems, facilitating the creation of a dataset for training improved attack models. The tool allows users to input images of perturbed license plates and generates output files that include annotated images and a CSV report detailing the effectiveness of various perturbations. Notable features include support for multiple YOLO detection and OCR models, along with a user-friendly interface to manage input and output data.

The algorithms tool provides minimal, well-documented implementations of various data structures and algorithms in Python 3, suitable for educational and practical applications. Its primary use case is to facilitate learning through code examples, covering areas like sorting, graph algorithms, dynamic programming, and data structures. Notable features include self-contained files with comprehensive docstrings, type hints, and complexity notes, making it an excellent resource for both developers and students.

AbletonMCP integrates Claude AI with Ableton Live via the Model Context Protocol, facilitating advanced prompt-assisted music production and live session manipulation. Key features include two-way communication, track and clip creation and editing, instrument and effect management, and autonomous song composition across Ableton's arrangement view. The tool operates through a Python-based MCP server and a MIDI Remote Script, ensuring seamless control over audio and MIDI elements within Ableton Live.

The 01 is an open-source voice interface platform designed for intelligent devices, enabling natural language communication with computers. Its primary use cases include executing code, web browsing, file management, and controlling third-party software, with notable features such as customization options for behavior and language models, and support for various server configurations to suit different hardware capabilities. The tool is currently in rapid development, emphasizing the importance of safety measures during its use.

YuE is an open-source series of foundation models for music generation, specializing in converting lyrics into complete songs. It features capabilities for generating diverse vocal tracks and accompaniment across various genres and languages, with support for incremental song generation, dual-track ICL mode, and a user-friendly Gradio interface for enhanced experimentation. Notable updates include LoRA finetuning, music continuation, and the ability to perform music style transfer, making it a versatile tool for creators in the music domain.

WouldYouKindly is a security testing tool designed to assess the resilience of large language models (LLMs) against simulated security attacks. It features customizable attack simulations utilizing natural language prompts, a dual agent setup for realistic defense scenarios, and automated scoring to evaluate an LLM’s performance in safeguarding sensitive information. This tool is particularly useful for developers and researchers aiming to enhance LLM security and robustness.

The WiFi Pineapple Pager Payload Library is a community-driven repository containing DuckyScript™ and Bash payloads specifically designed for the Hak5 WiFi Pineapple Pager. Its primary use case is to enable users to develop and share payloads for security testing and penetration testing, while notable features include a collaborative environment for submission and modification of payloads, along with a leaderboard for highlighting popular contributions.

websockets is a Python library designed for creating WebSocket servers and clients, emphasizing correctness, simplicity, robustness, and performance. Built on the asyncio framework, it offers a coroutine-based API, as well as support for threading and trio, making it suitable for various asynchronous programming needs. Notably, it ensures compliance with RFC 6455 and optimizes memory usage, making it a reliable choice for production environments.

WargamesAI is a comprehensive toolkit designed for the creation and simulation of wargames utilizing Large Language Model agents to enhance gameplay realism and decision-making. Its primary use case includes automating game design processes, managing gameplay dynamics through LLM umpiring, and facilitating participation by both LLMs and humans. Notable features encompass automated game scenario generation, integration of traditional game mechanics, and extraction of agent personalities from text documents, enabling rich, strategic simulations based on historical or fictional contexts.

VulnSage is an evaluation framework designed to benchmark Large Language Models (LLMs) in zero-shot software vulnerability detection (SVD) using a curated dataset of 593 real-world vulnerabilities categorized across 52 CWE types. It features multi-granular analysis capable of assessing vulnerabilities at function, file, and inter-function levels, alongside four distinct zero-shot prompting strategies for improved reasoning accuracy. Additionally, VulnSage employs a noise quantification methodology to enhance dataset reliability and evaluate LLM robustness against real-world vulnerabilities.