Metraipolis › archive › dc_3d348c842fd2

Does Showing Agents the Privilege of Their Tools Reduce Over-Privileged Tool Selection? — arXiv:2610.40975

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.40975
License (as archived)
CC0-1.0
Archived
2026-11-02T04:50:22.617Z
SHA-256 of the archived text
f565a5124d0e30415a8d072d65b1bbf0412656d3d2d24aef11a27d9d777be2f3
Stable URL of this page
https://metraipolis.com/documento/dc_3d348c842fd2

How to cite

Does Showing Agents the Privilege of Their Tools Reduce Over-Privileged Tool Selection? — arXiv:2610.40975 — arXiv (en), https://arxiv.org/abs/2610.40975 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3d348c842fd2 (Metraipolis archive dc_3d348c842fd2, sha256 f565a5124d0e30415a8d072d65b1bbf0412656d3d2d24aef11a27d9d777be2f3)
BibTeX
@misc{dc3d348c842fd2,
  title = {Does Showing Agents the Privilege of Their Tools Reduce Over-Privileged Tool Selection? — arXiv:2610.40975},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.40975},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3d348c842fd2; sha256 f565a5124d0e30415a8d072d65b1bbf0412656d3d2d24aef11a27d9d777be2f3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40975. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

When several tools can complete a task, an agent that picks the most powerful one holds more authority than the task needs. Recent work demonstrates that language-model agents do this often, and more so after a tool fails. Approaches to restrict focus so far on either enforcing least privilege outside the model or retraining the model. We ask a simpler question: does it help to tell the agent what each tool's privilege is? To investigate this question, we run all 544 ToolPrivBench scenarios on eight models, three run locally, and five through a hosted API, under five conditions: (1) no signal, (2) a least-privilege instruction, (3) privilege class and cost shown in each tool description, (4) both the instruction and privilege information, and (5) a matched neutral-metadata control. Showing privilege lowers premature escalation on every model that over-privileges at a measurable rate, reducing over-privileged selection by up to 19.3 percentage points (50\%), whereas irrelevant metadata in the same format does not. On most models, however, information alone mainly delays escalation until the standard tools have been tried. Combined with the instruction it more often avoids escalation, and the share of scenarios completed without a risk tool rises on seven of the eight models, reaching 51.1\% safe completion from a baseline of 18.8\% on the strongest-performing configuration. Visible privilege is therefore a cheap complement to enforcement rather than a substitute.

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/2610.40975 — 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 2026-11-02T04:44:42.666Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).