CoopIS-DT 2026 Abstracts


Area 1 - Ethical, Privacy-Preserving, and Secure Information Systems

Short Papers
Paper Nr: 9
Title:

SR-Sim: Reliability-Aware Remediation Simulation for Black-Box Security Ratings

Authors:

Nada Hanad, Mehdi Acheli, Ali Nour Eldin, Mohamed Sellami and Walid Gaaloul

Abstract: Commercial security rating platforms summarize an organization's externally observable cyber exposure as a numerical score used in procurement, cyber insurance, and third-party risk management. Organizations want to know not only their current score but which remediation actions would improve it and by how much. Yet, answering such what-if questions directly can reveal the platform's proprietary scoring logic. We demonstrate \toolname, an interactive tool that predicts the score impact of candidate remediation actions on the Web analysis domain of a commercial platform through a machine learning surrogate, without querying the engine. It attaches a calibrated reliability signal to every predicted impact, so untrustworthy predictions are flagged rather than shown as certain. The reliability layer is trained on observed remediation transitions and carries a selective-acceptance guarantee through Conformal Risk Control. A visitor composes a plan from Web-domain actions and reads the projected score with its confidence, shown either in an exact-point operator mode or in a customer mode that reveals only grade movement, preserving opacity.

Area 2 - Software Engineering for AI Engineering for Information Systems

Short Papers
Paper Nr: 5
Title:

Prompt Radar: An IDE Extension for Analyzing Prompt Smells in LLM-Enabled Software

Authors:

Vanilson Burégio and Ilyes Jenhani

Abstract: Large language model (LLM)-enabled applications increasingly rely on prompts embedded directly in source code. These prompts influence output quality, reliability, safety, security, response structure, and API cost, yet developers have limited support for inspecting them during development. This paper presents Prompt Radar, a Visual Studio Code extension that helps developers analyze prompts-as-code for potential prompt smells. Prompt Radar uses LLMs to assess prompts across five categories: Security, Safety, Reliability, Efficiency, and Format, and presents the results directly in the IDE through explanations and visual feedback. We evaluate Prompt Radar using real prompts mined from public GitHub repositories. A multi-LLM evaluation with ten models found weak agreement on exact smell scores but 92% agreement when scores differing by at most one level were considered a match. The results also show that Security is the most contested category, where developer review is especially important. To make the tool accessible, we provide a website with the GitHub repository, a usage demo video, and a survey for collecting feedback on usability, reliability, usefulness, and future improvements.

Paper Nr: 6
Title:

Demo: An AI-Augmented MBSE Toolchain for Ontology-Driven Digital Twin Interoperability

Authors:

Racem Bougacha and Slim Kallel

Abstract: We demonstrate an AI-augmented MBSE toolchain for gen- erating ontology-based representations of Digital Twin systems from natural- language specifications. The toolchain consists of three sequential mod- ules: (1) a LLaMA3-based natural-language-to-SysML generation mod- ule, (2) a rule-based SysML-to-OWL transformation module, and (3) an ontology matching and integration module. The demonstration uses two heterogeneous Digital Twin subsystems from the MAGPIE port logis- tics project, namely the Vessel Traffic Monitoring System (VTMS) and the Port Equipment Digital Maintenance System (PEDM). Attendees can follow the complete workflow from a textual specification to SysML models, generated OWL ontologies, ontology correspondences, and a uni- fied semantic model. The demonstration emphasizes transparency and inspectability by exposing intermediate artifacts at each stage. Quanti- tative results from the evaluated MAGPIE use cases include F1 = 0.77 for ontology matching. The paper focuses on the demonstrable realiza- tion and reproducibility of the workflow, while the detailed methodolog- ical evaluation is reported in our companion work [5]. Source code and demonstration resources are publicly available.

Paper Nr: 8
Title:

From Business Process Models to Runnable Applications by AI-Empowered AMADEOS

Authors:

Danijela Banjac, Zoran Djuric, Drazen Brdjanin, Mihajlo Savic and Goran Banjac

Abstract: Model-driven approaches to database design automate the derivation of conceptual database models from higher-level artifacts, but the resulting models are typically consumed only by technical users, and the generated output usually stops at the model level. This paper presents the most recent improvements to the AMADEOS tool for business process model-driven database design. In particular, it introduces a new client application centered around a project-based workflow and an AI-assisted feature employing large language models to augment reliable rule-based transformations. This latest feature transforms a conceptual model into a JHipster domain specification from which a complete modern web application can be generated.