aiqbench analysis
SkillGym introduces a novel framework for training LLM agents to effectively utilize skills by automatically generating verifiable environments and collecting high-quality trajectories. This approach significantly enhances an agent's ability to invoke relevant professional knowledge, demonstrating that targeted supervised fine-tuning can outperform much larger untrained models on complex, long-horizon tasks. The work addresses a critical gap in current agent paradigms by providing a scalable method for synthesizing reliable training data without relying solely on human annotation.
SkillGym: Training Skill-Use Agents with Automatic Verifiable Environment Generation
arXiv:2609.37539v1 Announce Type: new Abstract: Skills equip LLM agents with professional knowledge and guidance to complete long-horizon and complex tasks. Although skills have been widely adopted in recent agent paradigms and harnesses, how to synthesize reliable training data and how to train agents for skill use remain underexplored. In this work, we propose SkillGym, an automatic pipeline to build verifiable environments, collect trajectories, and train skill-use agents. SkillGym first crawls a large volume of skills from the internet, then keeps those whose workflows can run reproducibly offline. A builder-reviewer pipeline is used to construct difficulty-controlled tasks, spanning four task types, each with a reference solution and an executable verifier. With this pipeline, we build 6.8k environments and collect 19k verified successful trajectories for supervised finetuning. Finetuning on these trajectories improves LLMs of different families and sizes, from 2B to 122B parameters across four skill-use benchmarks; Our Qwen3.5-9B SFT model outperforms the 397B untrained model on two of them. Further analysis shows that training teaches agents to invoke skills, raising the rate of reading the relevant skill from 28% to 96%, and that the gains hold across reasoning structures, extending to task types that form a minority of the training data and to skills held out from training
Abstract:Skills equip LLM agents with professional knowledge and guidance to complete long-horizon and complex tasks. Although skills have been widely adopted in recent agent paradigms and harnesses, how to synthesize reliable training data and how to train agents for skill use remain underexplored. In this work, we propose SkillGym, an automatic pipeline to build verifiable environments, collect trajectories, and train skill-use agents. SkillGym first crawls a large volume of skills from the internet, then keeps those whose workflows can run reproducibly offline. A builder-reviewer pipeline is used to construct difficulty-controlled tasks, spanning four task types, each with a reference solution and an executable verifier. With this pipeline, we build 6.8k environments and collect 19k verified successful trajectories for supervised finetuning. Finetuning on these trajectories improves LLMs of different families and sizes, from 2B to 122B parameters across four skill-use benchmarks; Our Qwen3.5-9B SFT model outperforms the 397B untrained model on two of them. Further analysis shows that training teaches agents to invoke skills, raising the rate of reading the relevant skill from 28% to 96%, and that the gains hold across reasoning structures, extending to task types that form a minority of the training data and to skills held out from training
Artificial Intelligence (cs.AI)
arXiv:2609.37539 [cs.AI]
(or arXiv:2609.37539v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.37539 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 296 words
Related coverage
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · Jun 12, 2026
Run agent-driven Amazon SageMaker HyperPod operations with InstantStart
AWS · Sep 4, 2026
Echoverse: Deep, evolving environments for computer-use agents
Microsoft Research · Jul 30, 2026
Google AI Introduces EnvHarness: A Programmable Layer That Turns Static Agent Environments Into Adaptive Training Worlds
MarkTechPost · Aug 30, 2026