Metraipolis › archive › dc_7de530055f3a

Beyond Approved Actions: Runtime Validation of Persistent Outcomes in Agent Workflows — arXiv:2609.31301

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31301
License (as archived)
CC0-1.0
Archived
2026-10-08T01:39:51.295Z
SHA-256 of the archived text
69c9888c086db957035ee2621b9f4da29686926ae97fe86cdef464905491da75
Stable URL of this page
https://metraipolis.com/documento/dc_7de530055f3a

How to cite

Beyond Approved Actions: Runtime Validation of Persistent Outcomes in Agent Workflows — arXiv:2609.31301 — arXiv (en), https://arxiv.org/abs/2609.31301 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7de530055f3a (Metraipolis archive dc_7de530055f3a, sha256 69c9888c086db957035ee2621b9f4da29686926ae97fe86cdef464905491da75)
BibTeX
@misc{dc7de530055f3a,
  title = {Beyond Approved Actions: Runtime Validation of Persistent Outcomes in Agent Workflows — arXiv:2609.31301},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31301},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7de530055f3a; sha256 69c9888c086db957035ee2621b9f4da29686926ae97fe86cdef464905491da75},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31301. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model agents increasingly act on software systems, no longer merely generating text but also changing databases and online services. However, an approved database update may succeed yet leave an unapproved notification because execution can produce persistent effects beyond the requested change. Current safeguards can approve an action or record its aftermath, but without checking the persistent result before continuation, an unapproved outcome can be accepted as success and propagated to later steps. We present EffectMatch, a runtime that collects persistent changes within a controlled execution boundary and compares them with what the application approved for the current state and execution. The comparison governs commit and dependent execution. In comparative evaluation on 206 public business tasks, EffectMatch preserved all clean executions and prevented all tested incorrect commits. Six 20-run ablations exposed the failure caused by each removed mechanism, while 80 task-topology cases preserved truthful handoffs and blocked invalid continuation. Together, these results show that EffectMatch blocks the silent acceptance and downstream propagation of persistent outcomes inconsistent with application approval.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI).
Categorie arXiv: Software Engineering (cs.SE) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.31301 — 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 2026-10-08T01:12:14.571Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).