ResearchDatasetsClineFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsPrompt ExfiltrationCline coding agent prompts95.8Sim Score10Multi-tool Threshold Poisoning AttackCline Scenario II: Coding Assistant v3.35.096ASR4Multi-tool Threshold Poisoning AttackCline Scenario I Travel Assistant v3.35.096Attack Success Rate (ASR)4
Multi-tool Threshold Poisoning AttackCline Scenario I Travel Assistant v3.35.096Attack Success Rate (ASR)4