MSR 2026
Mon 13 - Tue 14 April 2026 Rio de Janeiro, Brazil
co-located with ICSE 2026
Mon 13 Apr 2026 11:50 - 12:00 at Oceania V - Session 1-A: AI Agents & Automation Chair(s): Matheus Paixao

In 2025, coding agents have seen a very rapid adoption. Coding agents leverage Large Language Models (LLMs) in ways that are markedly different from LLM-based code completion, making their study critical. Moreover, unlike LLM-based completion, coding agents leave visible traces in software repositories, enabling the use of MSR techniques to study their impact on SE practices. This paper documents the promises, perils, and heuristics that we have gathered from studying coding agent activity on GitHub.

Mon 13 Apr

Displayed time zone: Brasilia, Distrito Federal, Brazil change

11:00 - 12:30
Session 1-A: AI Agents & AutomationTechnical Papers / Industry Track / MSR Program at Oceania V
Chair(s): Matheus Paixao State University of Ceará
11:00
10m
Talk
Toward Linking Declined Proposals and Source Code: An Exploratory Study on the Go Repository
Technical Papers
Sota Nakashima Kyushu University, Masanari Kondo Kyushu University, Mahmoud Alfadel University of Calgary, Aly Ahmad University of Calgary, Toshihiro Nakae DENSO CORPORATION, Hidenori Matsuzaki DENSO CORPORATION, Yasutaka Kamei Kyushu University
Pre-print
11:10
10m
Talk
IntelliSA: An Intelligent Static Analyzer for IaC Security Smell Detection Using Symbolic Rules and Neural Inference
Technical Papers
Qiyue Mei The University of Melbourne, Michael Fu The University of Melbourne
Pre-print File Attached
11:20
10m
Talk
Model See, Model Do? Exposure-Aware Evaluation of Bug-vs-Fix Preference in Code LLMs
Technical Papers
Ali Al-Kaswan Delft University of Technology, Netherlands, Claudio Spiess University of California, Davis, Prem Devanbu University of California at Davis, Arie van Deursen TU Delft, Mali Izadi TU Delft
Pre-print
11:30
10m
Talk
A Match Made in Heaven? AI-driven Matching of Vulnerabilities and Security Unit Tests
Technical Papers
Emanuele Iannone Hamburg University of Technology, Quang-Cuong Bui Hamburg University of Technology, Riccardo Scandariato Hamburg University of Technology
Pre-print
11:40
10m
Talk
PhantomRun: Auto Repair of Compilation Errors in Embedded Open Source Software
Technical Papers
Han Fu , Sigrid Eldh Ericsson AB, Mälardalen University, Carleton University, Kristian Wiklund Ericsson AB, Andreas Ermedahl Ericsson AB; KTH Royal Institute of Technology, Philipp Haller KTH Royal Institute of Technology, Cyrille Artho KTH Royal Institute of Technology, Sweden
11:50
10m
Talk
Promises, Perils, and (Timely) Heuristics for Mining Coding Agent Activity
Technical Papers
Romain Robbes CNRS, LaBRI, University of Bordeaux, Théo Matricon Univ Rennes, INSA Rennes, Inria, CNRS, IRISA, Thomas Degueule CNRS, Andre Hora UFMG, Stefano Zacchiroli LTCI, Télécom Paris, Institut Polytechnique de Paris, Palaiseau, France
Pre-print Media Attached
12:00
10m
Talk
From Logic to Toolchains: An Empirical Study of Bugs in the TypeScript Ecosystem
Technical Papers
TianYi Tang Simon Fraser University, Saba Alimadadi Simon Fraser University, Nick Sumner Simon Fraser University
Pre-print
12:10
10m
Talk
Are We All Using Agents Now? An Empirical Study of Core and Peripheral Developers’ Use of Coding Agents
Technical Papers
Shamse Tasnim Cynthia University of Saskatchewan, Joy Krishan Das University of Saskatchewan, Banani Roy University of Saskatchewan
Pre-print
12:20
5m
Talk
Context Engineering for AI Agents in Open-Source Software
Technical Papers
Seyedmoein Mohsenimofidi Heidelberg University, Matthias Galster University of Canterbury, Christoph Treude Singapore Management University, Sebastian Baltes Heidelberg University
Pre-print
12:25
5m
Talk
A Blueprint for Trustworthy Code Annotation at Scale: An LLM-Powered Pipeline for Industrial Software Analytics
Industry Track