Work

Version 6.6.1 - AI-KM: Agent skills and ontology-driven knowledge modeling

Year: 2026

Type: software-paper

Abstract: This update introduces two significant capabilities: an agentic skill framework and ontology-driven knowledge modeling with retrieval. The agentic skill framework encapsulates complex AI behaviors as reusable, conversation-invokable skills, supporting both predefined expert skills and an autonomous planning mode for novel tasks. Meanwhile, the ontology-driven knowledge modeling module elevates knowledge representation from simple slice structures to formal, machine-interpretable ontologies built through natural language entity extraction. These advances enable domain experts to encode their expertise as both executable skills and structured knowledge, creating a closed loop where knowledge is represented as formal structure and applied as dynamic process.

Source: SoftwareX

DOI: 10.1016/j.softx.2026.102855

Authors: Haolin Wen, Songyi Wang, Bingjie Li, Junfen Xiong, Tong Chen, Xin Liang

Institutions: Naval University of Engineering, Southern University of Science and Technology, Peng Cheng Laboratory, General Hospital of Central Theater Command

Language: English

Topic: Semantic Web and Ontologies

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $1,200

Funder: National Social Science Fund of China