
StepJack
Benchmarking framework for evaluating computer-use AI agents against multi-step indirect prompt injection, with automatic adversarial goal…

Benchmarking framework for evaluating computer-use AI agents against multi-step indirect prompt injection, with automatic adversarial goal…

Proof-of-concept reproduction of an nginx heap overflow and info leak (CVE-2026-42533) with two attack surfaces, debug analysis, and a full RCE chain.

Set of tools to analyze Windows sandboxes for exposed attack surface.

CEREBRO-RED v2: Advanced LLM Red Team Research Platform with PAIR Algorithm and LLM-as-a-Judge Evaluation

Open-source web application security scanner for automated vulnerability detection, manual penetration testing, and API security testing with a…

Endpoint detection & Malware analysis software

Emulate and Dissect MSF and *other* attacks

Parses public sandbox detonation reports to produce threat hunting intelligence, organizes findings via MITRE ATT&CK, assembles IOCs, and generates…