Metraipolis › archive › dc_a1ddeada48aa
Runtime Authorization of Self-Generated Subgoals in Long-Horizon Tool-Using AI Agents — arXiv:2610.04975
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04975
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:24:55.391Z
- SHA-256 of the archived text
1c3d2b49e8212df01ab5378079763c8f9dc216f66709e411a1f4d79c6b644824- Stable URL of this page
https://metraipolis.com/documento/dc_a1ddeada48aa
How to cite
Runtime Authorization of Self-Generated Subgoals in Long-Horizon Tool-Using AI Agents — arXiv:2610.04975 — arXiv, https://arxiv.org/abs/2610.04975 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a1ddeada48aa (Metraipolis archive dc_a1ddeada48aa, sha256 1c3d2b49e8212df01ab5378079763c8f9dc216f66709e411a1f4d79c6b644824)
BibTeX
@misc{dca1ddeada48aa,
title = {Runtime Authorization of Self-Generated Subgoals in Long-Horizon Tool-Using AI Agents — arXiv:2610.04975},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04975},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a1ddeada48aa; sha256 1c3d2b49e8212df01ab5378079763c8f9dc216f66709e411a1f4d79c6b644824},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04975. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-horizon tool-using AI agents create subgoals, replan, delegate work, and compose sibling results. Per-tool permission checks cannot establish that a changing goal graph remains within the principal-approved task. We address this authorization gap in a finite structured domain with one principal and one authorization root. Each proposed goal-graph mutation carries a version-bound witness that its continuation traces, resources, obligations, invariants, and closing condition refine the active root contract; every protected effect is rechecked at an atomic commit boundary. Free-form goal text supplies no authority. We prove trace-policy and modeled forbidden-state preservation under explicit mediation, abstraction, freshness, and atomicity assumptions, plus conditional root-success preservation, a separation result for memoryless allowlists, exact finite-domain decidability, and universal-safety monotonicity under sound abstraction refinement. An executable model explores 340 states and 419 transitions. Across 96 matched cases covering 25 structural schemas, the complete mechanism commits zero forbidden states in 48 drifted cases and completes all 48 benign counterparts. Two public upstream runtime paths execute 258 native dispatches across 32 cases, with every case-level decision and receipt chain matching. A frozen host-local study covers 129 synthetic one-factor-at-a-time cells, all matching fixed decisions and reasons. A history-aware continuation comparator blocks every modeled bad trace prefix but commits all operations in 11 cases whose violations lie in typed resources, freshness, or explicit-join evidence outside its trace projection. Within the registered structured domains, runtime authorization preserves useful replanning while preventing self-generated subgoals from becoming a source of new authority. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2610.04975 — 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 2026-10-06T05:24:05.537Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).