ResearchTasksTool-invocation hijacking to RCETool-invocation hijacking to RCEFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAgent Security Scenarios Old LLM LineupRCE-2100RCE Success Rate20Jun 30, 2026Agent Security Scenarios New LLM LineupRCE-2100Gemini-3.1-pro RCE Success Rate5Jun 30, 2026