SkillNet: Create, Evaluate, and Connect AI Skills

TL;DR

SkillNet is an open infrastructure that systematically creates, evaluates, and organizes AI skills, significantly improving agent performance and enabling scalable skill transfer and consolidation.

Contribution

It introduces a unified ontology and platform for creating, evaluating, and connecting AI skills from diverse sources, facilitating scalable skill management.

Findings

Improves agent rewards by 40% on benchmark tasks.

Reduces execution steps by 30% across models.

Supports over 200,000 skills in the repository.

Abstract

Current AI agents can flexibly invoke tools and execute complex tasks, yet their long-term advancement is hindered by the lack of systematic accumulation and transfer of skills. Without a unified mechanism for skill consolidation, agents frequently ``reinvent the wheel'', rediscovering solutions in isolated contexts without leveraging prior strategies. To overcome this limitation, we introduce SkillNet, an open infrastructure designed to create, evaluate, and organize AI skills at scale. SkillNet structures skills within a unified ontology that supports creating skills from heterogeneous sources, establishing rich relational connections, and performing multi-dimensional evaluation across Safety, Completeness, Executability, Maintainability, and Cost-awareness. Our infrastructure integrates a repository of over 200,000 skills, an interactive platform, and a versatile Python toolkit. Experimental evaluations on ALFWorld, WebShop, and ScienceWorld demonstrate that SkillNet significantly enhances agent performance, improving average rewards by 40% and reducing execution steps by 30% across multiple backbone models. By formalizing skills as evolving, composable assets, SkillNet provides a robust foundation for agents to move from transient experience to durable mastery.