ResearchTasksAgentic Skill-UseAgentic Skill-UseFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAgentic Skill-Use Gold + Distr.Opus-4.780Final Success Rate7Jul 3, 2026Agentic Skill-Use Gold SkillsOpus-4.788Final Score7Jul 3, 2026Agentic Skill-Use No SkillsGPT-5.524Final Score7Jul 3, 2026