ResearchTasksRed-teaming tool-calling agentsRed-teaming tool-calling agentsFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMCP Filesystem 1.0 (test)Zero-Shot (ZS)68.8Success Rate (RR)5Mar 25, 2026MCP Playwright 1.0 (test)Zero-Shot (ZS)95.3RR5Mar 25, 2026MCP Gmail 1.0 (test)Zero-Shot (ZS)0.844RR5Mar 25, 2026MCP Slack 1.0 (test)Zero-Shot (ZS)90.6RR5Mar 25, 2026MCP CodeExecutor 1.0 (test)Zero-Shot (ZS)100RR5Mar 25, 2026