CapFence: Reducing Computational Overhead in Agentic AI via Selective Authenticated Delegation
International Journal of Creative and Open Research in Engineering and Management · 10.05.2026 · 02 (05) · 1-15
DOI: 10.55041/ijcope.v2i5.326 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Creative and Open Research in Engineering and Management, online erschienen am 10.05.2026. Verfasst von Anjali Srivastava Anjali Srivastava und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Agentic AI systems are increasingly ca-pable of executing multi-step plans, in-voking external tools, browsing the web, and modifying persistent state with lim-ited human supervision. This capability expansion introduces a critical yet un-deraddressed security challenge: by what authority does an agent act, and how can that authority be enforced efficiently at runtime? Current approaches to delega-tion safety are either static and brittle (allowlist-based access control), computa-tionally prohibitive (always-on large lan-guage model monitoring), or retrospec-tive and non-preventive (logging-only au-diting). None is suitable for the resource-constrained environments—edge devices, on-device copilots, bandwidth-limited en-terprise endpoints—where agentic sys-tems are rapidly being deployed. This paper proposes CapFence, a lightweight runtime delegation layer that addresses this gap through four tightly integrated components: (i) a capabil-ity compiler that translates natural-language intent into structured, least-privilege capability tokens; (ii) a com-pact risk gate that scores each pro-posed action using a low-cost calibrated model; (iii) a selective escalation mecha-nism that reserves expensive verification for genuinely ambiguous actions; and (iv) a compressed audit chain that sup-ports retrospective accountability with-out verbose trace retention. Motivated by the authenticated-delegation prob-lem articulated by South et al. [1] and the deployment-visibility tradeoffs doc-umented by Chan et al. [2], CapFence targets simultaneous reductions in de-cision latency (≈75%), token cost over-head (≈77%), false-positive rate (≈46%), and peak memory footprint relative to an always-on LLM monitor, while re-taining task success on the GAIA, We-bArena, Mind2Web, VisualWebArena, and AgentBench benchmark suites. The primary contribution is a governance primitive that makes least-privilege del-egation both principled and practically deployable. Keywords: Agentic AI, authenticated delegation, runtime gating, capability to-kens, selective escalation, auditability, edge deployment.
Zitieren
APA
Srivastava, A. S. A., Gupta, P. G. P., & Gulati, P. G. P. (2026). CapFence: Reducing Computational Overhead in Agentic AI via Selective Authenticated Delegation. International Journal of Creative and Open Research in Engineering and Management, 02(05), 1-15. https://doi.org/10.55041/ijcope.v2i5.326
Vancouver
Srivastava ASA, Gupta PGP, Gulati PGP. CapFence: Reducing Computational Overhead in Agentic AI via Selective Authenticated Delegation. International Journal of Creative and Open Research in Engineering and Management. 2026;02(05):1-15. doi:10.55041/ijcope.v2i5.326
BibTeX
@article{Srivastava2026,
author = {Srivastava, Anjali Srivastava Anjali and Gupta, Piyush Gupta Piyush and Gulati, Payal Gulati Payal},
title = {CapFence: Reducing Computational Overhead in Agentic AI via Selective Authenticated Delegation},
journal = {International Journal of Creative and Open Research in Engineering and Management},
year = {2026},
volume = {02},
number = {05},
pages = {1--15},
doi = {10.55041/ijcope.v2i5.326}
}
EUVIMED