SkillsBench evaluates how well agent skills work and how effectively agents use them across practical tasks.
unassessed
| Category | agentic |
|---|---|
| Subcategory | agent skill use |
| Page status | active |
| Metric | task success rate |
| Direction | higher_is_better |
| Unit | % |
| Publisher | BenchFlow AI |
Agent task success with and without selected skills, under the benchmark's task and tool-use protocol.
Agent interaction with task environments, tools and skill instructions.
No model card in ModelSpec reports this benchmark yet.