
StepJack
Benchmarking framework for evaluating computer-use AI agents against multi-step indirect prompt injection, with automatic adversarial goal…
adversarial-attackai-securitydynamic-analysis-sandboxing+2
6

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

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

Intel Pin-based tracer for API calls, syscalls, and instructions with anti-debug evasion, used for malware analysis and reverse engineering of packed…

Python antivirus evasion tool