Metraipolis › archive › dc_cddcf1f72c48

SkillAlchemy: Open-World Agent Skill Creation — arXiv:2608.23417

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.23417
License (as archived)
CC0-1.0
Archived
2026-10-18T07:41:57.074Z
SHA-256 of the archived text
19ae8867059cba76adc116b0e58b589d8b0bc88894a62751ef9ad2780cf7550f
Stable URL of this page
https://metraipolis.com/documento/dc_cddcf1f72c48

How to cite

SkillAlchemy: Open-World Agent Skill Creation — arXiv:2608.23417 — arXiv (en), https://arxiv.org/abs/2608.23417 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cddcf1f72c48 (Metraipolis archive dc_cddcf1f72c48, sha256 19ae8867059cba76adc116b0e58b589d8b0bc88894a62751ef9ad2780cf7550f)
BibTeX
@misc{dccddcf1f72c48,
  title = {SkillAlchemy: Open-World Agent Skill Creation — arXiv:2608.23417},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.23417},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cddcf1f72c48; sha256 19ae8867059cba76adc116b0e58b589d8b0bc88894a62751ef9ad2780cf7550f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.23417. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agent skills are reusable procedural artifacts that extend language agents with specialized workflows, tool conventions, and domain behaviors at inference time. However, creating reliable skills still depends largely on human authorship, model priors, or execution traces. These sources are often unavailable for unfamiliar tasks, suggesting the need to create skills from open-world materials. In this paper, we study open-world skill creation: given an underspecified skill brief and a source-access specification, a creator must discover behavior-relevant requirements omitted by the brief and determine how broadly each source-derived procedure is justified. We propose SkillAlchemy, an admission-centered framework for source-grounded skill creation. SkillAlchemy identifies implicit requirements through contrastive evidence, admits candidate procedures based on evidence-supported scope, and compiles the admitted content into a grammar-guided skill package. Extensive experiments across 87 SkillsBench v1.1 tasks demonstrate that SkillAlchemy improves pass rate over no-skill execution by 19.9 percentage points and the strongest automated baseline by 8.6 percentage points, while achieving performance comparable to human-curated skills.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.23417 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-18T07:37:52.779Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).