July 2026Unreviewed
AgenticRepair: Multi-Faceted Program Context Engineering for Agentic Vulnerability Repair
Michael Fu, Qiyue Mei, Patanamon Thongtanunam, Kla Tantithamthavorn
Abstract
Automated vulnerability repair aims to reduce the time and effort required to patch security flaws from a vulnerability triage report. Recent agentic AI approaches have shown promising results in automated program repair. However, vulnerability repair demands richer program context than general bug repair - context that security engineers routinely assemble in practice but that existing agentic approaches do not engineer. We identify three critical gaps: code-structure context capturing cross-fi
Categories
Cite
@misc{fu2026agenticrepair,
title = {{AgenticRepair: Multi-Faceted Program Context Engineering for Agentic Vulnerability Repair}},
author = {Michael Fu and Qiyue Mei and Patanamon Thongtanunam and Kla Tantithamthavorn},
year = {2026},
month = jul,
eprint = {2607.29422},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.29422}
}