Benchmark

Siren AgentDojo Utility

Siren AgentDojo evaluates tool-using agents under adversarial prompt-injection attacks. This metric reports utility on the assigned tasks.

Modality
text
Categories
safety, agents, tool_calling
Openness
unknown
Source
llm_stats
Reported scores
1

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Muse Glimmer-30BMeta0.9422026-08-10evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.