
TNC-Defense
Research code for detecting and detoxifying backdoors in text-to-image diffusion models, with pipelines for Stable Diffusion v1.4, v1.5, XL, and 3…

Research code for detecting and detoxifying backdoors in text-to-image diffusion models, with pipelines for Stable Diffusion v1.4, v1.5, XL, and 3…

Research code for poisoning attacks on the PGM-index, demonstrating how to craft adversarial data to degrade learned index performance.

Public version of the Entropica repo

AGLS

Agent Control Protocol (ACP) — Official English specification. Cryptographically verifiable authorization architecture for autonomous AI agents.

Code Implementation of "Pattern Enhanced Multi-Turn Jailbreaking: Exploiting Structural Vulnerabilities in Large Language Models"

Research code implementing backdoor attack and defense methods for LLMs, including IBSD, SLIP, BeDKD, and BadApex algorithms.

Code for ACL 2026 (main) paper "DeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation"

Research toolkit for analyzing AI agent behavioral patterns through multi-disciplinary corpus analysis. Parses session logs, runs 23 analytical…

AISec Plus Week 1 threat write-up — EchoLeak (CVE-2025-32711), zero-click indirect prompt injection in Microsoft 365 Copilot.

DugganUSA threat-intelligence contributions to the IETF Hackathon — real-world agentic-attack benchmark vectors, CVE-2026-33697 attestation analysis,…

Two-stage prompt-injection and jailbreak detector: regex gates plus a quantised DeBERTa-v3 ONNX classifier, with image, document, and audio support.…

An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning in Retrieval-Augmented Generation Systems.

Ensemble framework for software vulnerability detection and repair using multiple large language models, with consensus analysis and evaluation tools…

Multi-stage prompt injection technique that bypasses LLM safety alignment via identity reassignment, refusal suppression, and output coercion,…

Multi-layer security framework for AI agent ecosystems. Provides pre-installation skill auditing, file integrity monitoring, runtime protection, and…

The code in "DCVD: Dual-Channel Cross-Modal Fusion for Joint Vulnerability Detection and Localization"

Reference implementation of a multi-bit LLM watermarking scheme using coded payload spreading, unbiased reweighting, and soft-decision ECC decoding…