List your company

Vulnerability Attack Performance on E2E Cross-domain (dev)

0.8333ASR (E-commerce) (CodeQL)Gemini-3 Flash
0.3826680.4996590.616650.733641Feb 2, 2026

Evaluation results

MethodLinks
Gemini-3 FlashEvaluation Protocol=Cross-domain2026.020.83330.66670.83330.666710.714310.71430.750.58330.750.58330.66670.63640.66670.63640.750.57140.750.5714
GPT-5.2Evaluation Protocol=Cross-domain2026.020.750.63640.750.63640.66670.66670.66670.666710.510.50.66670.50.66670.50.66670.60.66670.6
GrokEvaluation Protocol=Cross-domain2026.020.750.72730.750.72730.60.70.60.70.750.66670.750.66670.66670.66670.66670.66670.750.66670.750.6667
Gemini-3 ProEvaluation Protocol=Cross-domain2026.020.71430.72730.71430.727310.888910.88890.750.70.750.70.71430.750.71430.750.750.70.750.7
ComposerEvaluation Protocol=Cross-domain2026.020.50.54550.50.54550.66670.66670.66670.66670.60.70.60.70.50.57140.50.57140.60.50.60.5
Claude-4.5 OpusEvaluation Protocol=Cross-domain2026.020.40.90910.40.90910.50.77780.50.77780.60.77780.60.77780.40.77780.40.77780.60.90.60.9