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

Agent Task Success on tau2-bench Retail Domain

61.4Total Pass Rate

SkillsInjector

39.66445.30750.9556.593May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
61.4---
60---
59.8---
55.3---
55.1---
2026.05
54.9---
53---
2026.05
51.2---
40.5---