Metraipolis › archive › dc_0a30bfb0c244

Incomparable Minima in Agent Tool Authorization — arXiv:2610.31914

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.31914
License (as archived)
CC0-1.0
Archived
2026-10-26T20:07:08.632Z
SHA-256 of the archived text
08cfff704e48b9c73a237162a0e02de4b7a74b490471c55141c46b7211e280dc
Stable URL of this page
https://metraipolis.com/documento/dc_0a30bfb0c244

How to cite

Incomparable Minima in Agent Tool Authorization — arXiv:2610.31914 — arXiv (en), https://arxiv.org/abs/2610.31914 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0a30bfb0c244 (Metraipolis archive dc_0a30bfb0c244, sha256 08cfff704e48b9c73a237162a0e02de4b7a74b490471c55141c46b7211e280dc)
BibTeX
@misc{dc0a30bfb0c244,
  title = {Incomparable Minima in Agent Tool Authorization — arXiv:2610.31914},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31914},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0a30bfb0c244; sha256 08cfff704e48b9c73a237162a0e02de4b7a74b490471c55141c46b7211e280dc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31914. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

An agent that needs a tool needs authority to call it, and every proposal for governing that authority has to answer one question: of two grants that both let the agent finish a task, which one is smaller? Over-privilege benchmarks score agents against this answer, and permission solvers compute it. We check the question against the authorisation tables that Microsoft and Google publish, covering 22,305 Microsoft Graph endpoint bindings and 23,852 Google API methods, on tasks generated from each table's own resource structure. For 29.8-33.3% of Graph tasks and 3.2% of Google resource tasks the question has no answer, because the minimal sufficient grants form an antichain, a set of grants none of which authorises a subset of what another does. In its two main authorisation modes, Microsoft's shipped privilegeLevel rates 67.4-74.4% of these incomparable pairs equal. Google ships no such scale, so on Google we apply a reconstruction of the FORTIS privilege ladder, which rates 90.7-98.7% equal. Applying the same reconstruction to Graph attributes much of that difference to the scale rather than the corpus: its form based on the HTTP verb alone rates 92.0% of Google pairs and 89.0-92.0% of Graph pairs equal. Four risk measures (operations, writes, services reached, administrative surface) order 28.7% of incomparable Google pairs and 15.7-23.4% of Graph pairs in opposite directions, so for those pairs the ranking has no measure-independent answer. Offered pools of minimal and larger grants, 6 frontier models pick a minimal grant on 99.0% of 1,708 decisions, a grant smallest under all four measures on 66.7%, and on 22.1% a grant that another option matches or beats on every measure. We propose that benchmarks report which minimal grant an agent chose and that deployments show the alternatives to whoever approves the grant.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2610.31914 — 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 2026-10-26T19:48:33.250Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).