Metraipolis › archive › dc_cea2ebd66ef6

Agent Skill Evolution: How Revisions Affect Coding Agents — arXiv:2610.04832

Source
arXiv
Original
https://arxiv.org/abs/2610.04832
License (as archived)
CC0-1.0
Archived
2026-10-06T05:32:40.057Z
SHA-256 of the archived text
98ba18601aae84239cc8ca8167651c12ee3082c6f3cad8d310be46026787d770
Stable URL of this page
https://metraipolis.com/documento/dc_cea2ebd66ef6

How to cite

Agent Skill Evolution: How Revisions Affect Coding Agents — arXiv:2610.04832 — arXiv, https://arxiv.org/abs/2610.04832 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cea2ebd66ef6 (Metraipolis archive dc_cea2ebd66ef6, sha256 98ba18601aae84239cc8ca8167651c12ee3082c6f3cad8d310be46026787d770)
BibTeX
@misc{dccea2ebd66ef6,
  title = {Agent Skill Evolution: How Revisions Affect Coding Agents — arXiv:2610.04832},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04832},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cea2ebd66ef6; sha256 98ba18601aae84239cc8ca8167651c12ee3082c6f3cad8d310be46026787d770},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04832. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agent Skills, the this http URL files that tell an LLM coding agent how a project works, are revised like code, yet what a revision does to the agent is unknown. From 2,608 first/last revision pairs of 3,159 Skills, we characterize how Skills evolve and how they change together with the configuration of the agent's harness. We then focus on rule changes, revisions that add or remove a rule we can check automatically, such as "run allium check". We measure their effect on 21 models in single answers and on four agents in a sandbox, and their cost on 20 of these models and the four agents. Most revisions (55%) change a rule or procedure, and commits that revise a Skill change harness files such as this http URL more often than other commits of the same size. Across 16 open-weight models, an added rule raises compliance in a single answer by +0.41 on average. Across the four agents, the rate at which the agent takes the required action rises by +0.23 on average (+0.16 to +0.36), and for the three agents that blind judges assessed, final correctness rises by +0.10 on average (+0.06 to +0.14). The gain comes mainly from rules that name a command or path the old Skill did not mention. Real tools load a Skill's body only when the agent decides it needs it. In that setting the four agents keep about half of the action gain on average (51%), and the three open models about 38%. A revision adds 18-19% input tokens to a single answer and no detectable cost to an agent episode, while loading a Skill's body raises the tokens of an episode by 50% on average.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.SE, nuovi lavori del giorno).
Categorie arXiv: Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2610.04832 — 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 2026-10-06T05:27:20.394Z dalla pagina https://arxiv.org/list/cs.SE/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).