PromptSentinel-X: A Leakage-Aware and Context-Aware Framework for Prompt-Injection Detection in Large Language Model-Powered Web Agents
Future Internet · 19.07.2026 · 18 (7) · 376
DOI: 10.3390/fi18070376 ↗Kurz gefasst
Fachartikel in der Zeitschrift Future Internet, online erschienen am 19.07.2026. Verfasst von Lily Popova Zhuhadar. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Large language model (LLM)-powered web agents combine privileged instructions with user requests, webpages, retrieved documents, tool outputs, memory, and conversation history, creating prompt-injection risks that static text classification may not capture. This study presents PromptSentinel-X, a leakage-aware and context-aware screening framework. The primary benchmark contained 1581 English records from the Prompt Injection Malignant dataset; 30,015 external prompt-injection, jailbreak, and benign hard-negative records were reserved for transfer and stress testing. The framework uses prompt-family-aware partitioning, trusted–untrusted context segmentation, calibrated risk prediction, robustness analysis, and deployment-oriented routing. On a 465-record group-aware test set, PromptSentinel-X achieved 0.9849 accuracy, 0.8887 macro-F1, 0.9851 attack recall, a 0.0050 benign false-positive rate, 0.9971 area under the receiver operating characteristic curve (AUROC), 0.9836 area under the precision-recall curve (AUPRC), and 0.0153 expected calibration error. Random splitting produced a higher baseline macro-F1 but 144 leakage warnings. Context-aware macro-F1 declined from 0.8301 for static prompts to 0.6306 for multi-turn, 0.3844 for retrieval-augmented generation, and 0.3200 for memory scenarios. PromptSentinel-X is positioned as a calibrated screening and escalation component, not a stand-alone authorization mechanism. Broader tool, memory, multimodal browser, multilingual, and end-to-end agent studies remain necessary.
Zitieren
APA
Zhuhadar, L. P. (2026). PromptSentinel-X: A Leakage-Aware and Context-Aware Framework for Prompt-Injection Detection in Large Language Model-Powered Web Agents. Future Internet, 18(7), 376. https://doi.org/10.3390/fi18070376
Vancouver
Zhuhadar LP. PromptSentinel-X: A Leakage-Aware and Context-Aware Framework for Prompt-Injection Detection in Large Language Model-Powered Web Agents. Future Internet. 2026;18(7):376. doi:10.3390/fi18070376
BibTeX
@article{Zhuhadar2026,
author = {Zhuhadar, Lily Popova},
title = {PromptSentinel-X: A Leakage-Aware and Context-Aware Framework for Prompt-Injection Detection in Large Language Model-Powered Web Agents},
journal = {Future Internet},
year = {2026},
volume = {18},
number = {7},
pages = {376},
doi = {10.3390/fi18070376}
}
EUVIMED