Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Agent Task Completion Average on tau2-bench, SkillsBench, and ALFWorld

58.7Average Pass Rate

SkillsInjector

23.44432.59741.7550.903May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
58.7
53.6
53.5
49.7
48.5
2026.05
46.7
42.6
2026.05
40.2
24.8