Progressive Agent Skill Generation via Reinforcement Learning
Open paper
Benchmark
Model
Rank
Results
agents-on-bench-airline
Claude Sonnet 4.5 + Skill-α
#2
Accuracy: 80.0
agents-on-bench-retail
Claude Sonnet 4.5 + Skill-α
#2
Accuracy: 86.0
Rank counts only results with a code link.