Metraipolis › archive › dc_7658b6db8fcb

Certified Execution Harness for Agentic AI: Safe and Least-Intrusive Petri-Net Supervisory Control Between Reasoning and Action — arXiv:2610.31376

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.31376
License (as archived)
CC0-1.0
Archived
2026-10-25T08:33:33.360Z
SHA-256 of the archived text
98de8d9da5a28910242e7ec7000bdfa5dec1ae3d19157e4aa3e873ed10055da3
Stable URL of this page
https://metraipolis.com/documento/dc_7658b6db8fcb

How to cite

Certified Execution Harness for Agentic AI: Safe and Least-Intrusive Petri-Net Supervisory Control Between Reasoning and Action — arXiv:2610.31376 — arXiv (en), https://arxiv.org/abs/2610.31376 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7658b6db8fcb (Metraipolis archive dc_7658b6db8fcb, sha256 98de8d9da5a28910242e7ec7000bdfa5dec1ae3d19157e4aa3e873ed10055da3)
BibTeX
@misc{dc7658b6db8fcb,
  title = {Certified Execution Harness for Agentic AI: Safe and Least-Intrusive Petri-Net Supervisory Control Between Reasoning and Action — arXiv:2610.31376},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.31376},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7658b6db8fcb; sha256 98de8d9da5a28910242e7ec7000bdfa5dec1ae3d19157e4aa3e873ed10055da3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31376. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This paper studies the problem of safe tool execution for large language model (LLM) agents under prompt injection. Using Petri nets, we formalize the execution harness between the LLM and its tools as a supervisory control problem, in which marks represent capabilities, approvals and budgets. Since a transition fires only if the marks it needs are present, an injected request whose preconditions are unmet or already spent is refused regardless of its wording. Treating tool outcomes as uncontrollable, we enforce safety by the conventional maximally permissive supervisor, and this safety transfers to the running system under four assumptions. Moreover, to capture an agent that may propose again after a refusal, we introduce the notion of a replanning closed loop, in which each retry costs an inference call. We show that, in this loop, the conventional supervisor can be arbitrarily more intrusive than a stricter safe one. Furthermore, for a given proposer model, the maximally permissive least-intrusive supervisor is characterized as a value-threshold refinement of the conventional one. Under precise conditions, this refinement orders two concurrent actions by at most one monitor place. On an autonomous release agent, we find by synthesis that safe deployment requires a step that the supervisor can refuse after a rollout starts.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Systems and Control (eess.SY).
Pagina del lavoro: https://arxiv.org/abs/2610.31376 — 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 2026-10-25T08:31:13.090Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).