Projects & Contributions

Some of the projects listed here are available on GitHub under the Apache 2.0 license, while others can be provided upon request.

ExoMind

Portable knowledge runtime for OpenClaw + Codex, combining markdown-first memory with associative retrieval. Supports indexing, recall, benchmarking, and automation for local-first personal knowledge workflows.

Apache 2.0

Abyss

AI-powered AppSec analysis platform for local-first security workflows with Local LLM support and isolated scanning. Includes project-based scanning, vulnerability triage, and practical remediation workflows for developers and security teams.

Apache 2.0

ArcStrike

Offensive security project focused on practical red-team workflows and operator-first execution. Designed for fast iteration, field utility, and realistic adversarial simulation in controlled environments.

Proprietary

Biometric Identification through Auditory EEG Signatures - YSC 2025

Developed secure EEG-based biometric identification system using auditory stimulus-induced brain signals with Deep Metric Learning techniques. Implemented deep learning models with metric learning approaches on PhysioNet Auditory EEG dataset, incorporating signal preprocessing and feature enhancement methods. System achieves competitive performance in individual classification and includes novel subject detection capability for enhanced real-world security applications.

Apache 2.0

Malware Detection using ML/DL - ATC 2024

First Author research presented at ATC 2024 Conference. Investigated machine learning and deep learning techniques for malware detection based on Windows PE files, comparing performance of classical ML and deep text models.

Apache 2.0

Deep Learning-based Intrusion Detection System

Developed advanced intrusion detection model using LSTM and Transformer architectures to detect web exploitation attacks. Achieved 94% accuracy with low false positive rates using ECML/PKDD 2007 and CSIC 2010 datasets.

Apache 2.0

Smart Contract Vulnerability Detection System

Built ML & DL-based vulnerability detection system for smart contracts. Trained on self-created Solidity contract datasets using SVC, XGBoost, and biLSTM models. Features interactive Streamlit demo for real-time vulnerability assessment.

Apache 2.0

SilentWebWeaver

Designed a C++ shellcode dropper/loader using HTTP protocol. Successfully bypassed Windows Defender and Kaspersky antivirus. Implemented advanced persistence techniques and encryption for secure payload delivery in Red Team operations.

Apache 2.0

Elegant ICMP Tunneling

Created a Python-based multi-client remote access tool over ICMP protocol. Managed secure communication channels and stored victim metadata in a custom database. Enables covert command and control operations through firewall-restricted networks.

Apache 2.0

CryptoWallet

Developed a secure cryptocurrency wallet with advanced security features including multi-signature support, hardware wallet integration, and encrypted key storage. Implemented robust protection against common attack vectors like phishing, malware, and man-in-the-middle attacks. Features secure transaction signing, cold storage capabilities, and comprehensive audit logging for blockchain security research.

Apache 2.0

© Yasna. All Rights Reserved.