List your company

Vulnerability Attack Analysis on WebGenBench E-commerce target-domain and cross-domain 1.0

100Attack Success Rate (CodeQL)GPT-5.2
4861.57588.5Feb 2, 2026

Evaluation results

MethodLinks
GPT-5.2Evaluation Protocol=Held-out (target-domain)2026.0210010010086.67
Claude-4.5 OpusEvaluation Protocol=Held-out (target-domain)2026.02100100100100
Gemini-3 ProEvaluation Protocol=Held-out (target-domain)2026.02100100100100
Gemini-3 FlashEvaluation Protocol=Held-out (target-domain)2026.02100100100100
ComposerEvaluation Protocol=Held-out (target-domain)2026.02100100100100
GrokEvaluation Protocol=Held-out (target-domain)2026.0210010010081.82
GPT-5.2Evaluation Protocol=Cross-domain2026.0276.4781.2574.2370.97
Claude-4.5 OpusEvaluation Protocol=Cross-domain2026.0272.7381.2572.7380
ComposerEvaluation Protocol=Cross-domain2026.0268.7576.1958.3375
GrokEvaluation Protocol=Cross-domain2026.0268.7566.676560.71
Gemini-3 ProEvaluation Protocol=Cross-domain2026.0261.547561.5465.52
Gemini-3 FlashEvaluation Protocol=Cross-domain2026.025066.675056.25