List your company

Agent Task Completion on tau2-bench airline

60Pass RateSkillsInjector
22.97632.58842.251.812May 28, 2026

Evaluation results

MethodLinks
SkillsInjector2026.0560
Graph of Skills2026.0556.1
SkillRouter2026.0554
LLM-as-selector2026.0549.6
Dense Cosine2026.0545.2
BM252026.0543.6
Random-skill2026.0542.4
No-skill2026.0537.6
Full-library2026.0524.4