backdoor-detection

Server Scanner detect rootkit and backdoor

Official repository for CVPR'23 paper: Detecting Backdoors in Pre-trained Encoders

Generate deterministic semantic fingerprints for Go functions. Detect xz-style backdoors and verify semantic equivalence using Static Single Assignment (SSA) analysis and Scalar Evolution.

BackdoorMBTI is an open source project expanding the unimodal backdoor learning to a multimodal context. We hope that BackdoorMBTI can facilitate the analysis and development of backdoor defense methods within a multimodal context.

Deterministic, read-only static auditor for GGUF models: behavioral chat-template backdoors + SSTI, tokenizer, config, and model-card surfaces - never renders, never loads weights. 192,032 repos, 137,698 templates analyzed, 0 reviewed false-positive FAILs.

This repository is the official implementation of the paper "ASSET: Robust Backdoor Data Detection Across a Multiplicity of Deep Learning Paradigms." ASSET achieves state-of-the-art reliability in detecting poisoned samples in end-to-end supervised learning/ self-supervised learning/ transfer learning.

[ECCV'24] T2IShield: Defending Against Backdoors on Text-to-Image Diffusion Models

[CCS'24] Official Implementation of "Fisher Information guided Purification against Backdoor Attacks"

[TPAMI 2025] Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models

Unofficial implementation of the detector cleanse in the paper "BadDet: Backdoor Attacks on Object Detection."

Solution for the Third Challenge of the Main Phase in the Rayan International AI Contest: Backdoored Model Detection.

Official PyTorch implementation of "Scanning Trojaned Models Using Out-of-Distribution Samples" (NeurIPS 2024)

Find the money dupe, not just the malware. A Claude Code skill that reviews FiveM and RedM resources for dupes, backdoors, crash vectors and supply-chain risk. Legacy + GTA V Enhanced, ESX/QBCore/QBox/ox.

Educational research demonstrating weight manipulation attacks in SafeTensors models. Proves format validation alone is insufficient for AI model security.

ALIRT(Advanced Linux incident response toolkit, formerly known as AIRT) is a Linux-based tool suite for incident response, specifically designed to detect kernel rootkits on the Linux box.

Rubber Ducky is an awareness initiative focused on educating individuals and organizations about potential cybersecurity threats that can be delivered through USB devices, particularly targeting issues like ransomware, backdoors, and keyloggers. Leveraging the concept of Bad USB and the functionality of Rubber Ducky USBs.

내 공유기가 Zbtlink ENDLESSDOORS 백도어(CVE-2026-66747) 대상인지 클릭 한 번으로 검사하는 Windows 프로그램

Runtime dependency-behavior monitor for Node.js. Two engines: in-process telemetry + an out-of-process (strace) trust boundary that sees native egress & persistence. Defense-in-depth for npm supply-chain attacks — SARIF, GitHub Action, zero deps.

weightwatch: detector black-box de backdoors en modelos open-weight fine-tuneados. GAP verificado 2026-08-27 (0/0/1 repos vs 75 papers arXiv:2608.11348). AGPL-3.0-or-later.

RKHUNTER LIVE is an immersive, interactive training platform for learning rootkit detection and malware forensics on Linux systems. Featuring a fully simulated rkhunter, chkrootkit, AIDE, and Lynis environment, this platform allows security professionals and students to practice identifying kernel rootkits, rootkits, userland rootkits🕵🏿.