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

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

A privacy-first app that strips AI watermarks from content you own.

A tool uses Windows Filtering Platform (WFP) to block Endpoint Detection and Response (EDR) agents from reporting security events to the server.

A tool for generating .NET serialized gadgets that can trigger .NET assembly load/execution when deserialized using BinaryFormatter from JS/VBS/VBA…

Dynamically invoke arbitrary unmanaged code from managed code without PInvoke.

Leverage a legitimate WFP callout driver to prevent EDR agents from sending telemetry

PoC code from DEF CON 25 presentation

Remove API hooks from a Beacon process.

Bypass Userland EDR hooks by Loading Reflective Ntdll in memory from a remote server based on Windows ReleaseID to avoid opening a handle to ntdll ,…

Data from a BRAWL Automated Adversary Emulation Exercise

A tool to find folders excluded from AV real-time scanning using a time oracle


A Poc on blocking Procmon from monitoring network events

Research code reproducing multi-turn LLM jailbreak experiments (FITD, MRCJ, ActorAttack, X-Teaming) from the SoK intent-oriented systematization…

Adversarial image attacks on vision-language web agents, from visual grounding to browser execution

A diagnostic framework for measuring LLM vulnerability to Affective Contextual Erosion (ACE) and related liminal attack vectors. **Delirium** is not…

Research artifact from USENIX Security 2026 demonstrating DDoS attacks by abusing tunnel nodes and IPv6 extension headers, with accompanying analysis…

Code and results for the QAI 2026 paper on FGSM attack transferability from simulation to IBM quantum hardware.