Metraipolis › archive › dc_ae61ba1b6f48

"Shall We Approve?" Consequence-Aware Approval for Autonomous Coding Agents — arXiv:2610.40753

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.40753
License (as archived): CC0-1.0
Archived: 2026-11-03T07:32:30.937Z
SHA-256 of the archived text: b05b45a4523bc606499b2d5a4cb2a3dacb55ee13815b4b6a47468ef6e1310e40
Stable URL of this page: https://metraipolis.com/documento/dc_ae61ba1b6f48

How to cite

"Shall We Approve?" Consequence-Aware Approval for Autonomous Coding Agents — arXiv:2610.40753 — arXiv (en), https://arxiv.org/abs/2610.40753 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ae61ba1b6f48 (Metraipolis archive dc_ae61ba1b6f48, sha256 b05b45a4523bc606499b2d5a4cb2a3dacb55ee13815b4b6a47468ef6e1310e40)

BibTeX

@misc{dcae61ba1b6f48,
  title = {"Shall We Approve?" Consequence-Aware Approval for Autonomous Coding Agents — arXiv:2610.40753},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.40753},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ae61ba1b6f48; sha256 b05b45a4523bc606499b2d5a4cb2a3dacb55ee13815b4b6a47468ef6e1310e40},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40753. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Coding agents increasingly rely on automated approvers to decide whether a proposed shell command may run. Existing approvers, from rule-based checkers to guardrail models and deployed permission classifiers, base their decisions on the text of the request and command. Yet a command's behavior depends on the code it invokes and the state of the workspace, and systems that observe file changes leave the assessment to the agent or a human after execution. We present Caper, a consequence-aware approval mechanism that pre-executes each command in a snapshotted sandbox, provides an independent approver with a bounded report of its observed effects, commits allowed commands, and rolls back blocked ones with a reason the agent can act on. Compared with six existing systems, Caper achieves the lowest harmful-outcome rate on SABER (6.7% versus at least 14.8%) and YoloFS-Ext, a benchmark we build in which harmful behavior is hidden inside invoked tools (7.8% versus at least 35.3%), while blocking 1.0% of the safe twins of the dangerous tasks in YoloFS-Ext.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.40753 — 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 2026-11-03T07:29:33.686Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).