ResearchBenchmarksVulnerability Detection on Projects P1-P8 (test)Vulnerability Detection on Projects P1-P8 (test)Follow17True Positives (TP)VibeGuard16.1516.5751717.425Apr 1, 2026Evaluation resultsMethodMethodLinksTrue Positives (TP)False Positives (FP)False Negatives (FN)True Negatives (TN)PrecisionRecallF1 ScoreGate AccuracyVibeGuardPolicy=DefaultPolicy=Default2026.0417202189.4710094.44100