Paper 2603.04448
SkillNet: Create, Evaluate, and Connect AI Skills
- Published
- Feb 2026
- Research lab
- Independent
- Citations
- 34
- GitHub
- 1.1K stars
01 In brief
Summary
SkillNet is an open infrastructure for creating, evaluating, and organizing AI skills at scale, addressing the lack of systematic skill consolidation in current AI agents.
It structures skills within a three-layer ontology (taxonomy, relation graph, package library) and supports automated skill creation from heterogeneous sources like execution trajectories, GitHub repositories, and documents.
A multi-dimensional evaluation framework assesses Safety, Completeness, Executability, Maintainability, and Cost-awareness, validated by human-LLM agreement (QWK near 1.0).
The repository contains over 200,000 skills (150,000+ curated) and includes a Python toolkit and API.
Experiments on ALFWorld, WebShop, and ScienceWorld show SkillNet improves average rewards by 40% and reduces execution steps by 30% across backbone models (DeepSeek V3.2, Gemini 2.5 Pro, o4 Mini).
Applications include autonomous scientific discovery, coding agents, and integration with OpenClaw.
Limitations include incomplete skill coverage, potential quality issues, and lack of an end-to-end pipeline.
02 From the paper
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.