Search papers, labs, and topics across Lattice.
This study empirically investigates AI agent skills as engineered artifacts, analyzing 18,463 skills from public registries and 23,199 from personal repositories to understand their authorship, reuse, customization, and maintenance. The findings reveal that while skills are predominantly reused in a near-verbatim manner, with 53% remaining unmodified post-adoption, the primary focus of modifications is on adapting operational specifications and local knowledge. The research highlights the importance of maintaining a stable behavioral contract across skills, suggesting that future efforts should prioritize project-specific adaptations and improved support for consolidating domain knowledge.
Most AI agent skills are reused verbatim, with over half never modified, revealing a critical gap in how we think about skill evolution and maintenance.
AI coding agents increasingly rely on skills: structured context bundles, typically a SKILL.md file with a YAML header and Markdown body, loaded on demand for domain knowledge, workflows, and scripts. Public registries such as skills.sh now host tens of thousands of skills, making them an emerging unit of reuse in agent-based software engineering. Yet skills have largely been viewed as agent capabilities rather than software artefacts whose content and evolution shape agent behaviour. We present the first empirical study of AI agent skills as engineered artefacts that are authored, reused, customised and maintained, across public registries and personal-use repositories. We mined 18,463 skills from skills.sh and 23,199 personal-use skills from 5,876 GitHub repositories, identifying 3,709 reuse links. LLM-based classification into SWEBOK knowledge areas (KAs) shows Software Construction dominates alongside a long tail of specialised areas. A thematic analysis of 180 skills identifies six content categories. Qualitative coding of 444 modifications reveals six themes, of which reworking operational specifications and adapting knowledge and resources are the primary target of change. Our findings show that reuse is largely a one-time copy operation: most reused skills remain near-verbatim, 53\% are never modified after adoption, and subsequent local maintenance is overwhelmingly additive. Customisation primarily adapts skills to local environments, whereas evolution accretes new inline domain knowledge. Across both, a stable behavioural contract -- how a skill interacts with users, monitors runtime state, and recovers from failures -- remains almost untouched. These results suggest maintenance effort should focus on project-specific bindings, and that registries and tool support should enable consolidating the domain knowledge skills re-author in isolation.
