EUVIMEDEUVIMED entdecken
Crossref

SECUREPATCH: Specialized Multi-Agent Architecture with Automated Validation for Security Vulnerability Repair

L. Yashwanth Reddy, M. N. S. Sumanth, R. Regin, S. Rubin Bose, S. Sharan Jeev, S. Benitta Sherine

AVE Trends in Intelligent Computing Systems · 2026

DOI: 10.64091/atics.2026.000283 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AVE Trends in Intelligent Computing Systems, 2026. Verfasst von L. Yashwanth Reddy und 5 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Software security flaws are crucial. Recent Large Language Model (LLM) techniques achieve 47-72% program repair success yet require specialized validation and iterative refinement. SECUREPATCH, a multi-agent LLM system with four specialized agents—Auditor (vulnerability detection), Architect (patch generation), Validator (security verification), and Coordinator (workflow management), achieves 99.4% security vulnerability repair success through iterative refinement with automated feedback. Researchers test three iteration algorithms (MAX RETRIES = 1, 3, 5) on 160 Python security vulnerabilities from 8 CWE categories. Our single-iteration setup (SECUREPATCH-1) outperforms all known approaches (47-72%) by 13-38%. Three cycles (SECUREPATCH-3) increase success to 95.6% (+10.6%), exceeding the 95% production reliability requirement. Five iterations (SECUREPATCH-5) enhance 99.4% (+14.4% overall) with 1 failure in 160 cases. Marginal analysis shows diminishing returns: 1→3 iteration transition (+10.6%) and 3→5 transition (+3.8%). Per-category analysis with five iterations yields 100% accuracy on 7 of 8 CWE categories. Weak Cryptography (CWE-327) improves the most (+45%, from 50% to 95%), showing iteration works on difficult security patterns. SECUREPATCH surpasses LLM-based approaches by 27-52% and typical automated program repair by 44-88%. Multiple-agent specialisation provides a strong baseline performance (85%); iterative refinement with validation feedback is essential for near-perfect accuracy (99.4%); and three iterations provide an optimal cost-benefit trade-off (95.6% success with 15% computational overhead).

Zitieren

APA

Reddy, L. Y., Sumanth, M. N. S., Regin, R., Bose, S. R., Jeev, S. S., & Sherine, S. B. (2026). SECUREPATCH: Specialized Multi-Agent Architecture with Automated Validation for Security Vulnerability Repair. AVE Trends in Intelligent Computing Systems. https://doi.org/10.64091/atics.2026.000283

Vancouver

Reddy LY, Sumanth MNS, Regin R, Bose SR, Jeev SS, Sherine SB. SECUREPATCH: Specialized Multi-Agent Architecture with Automated Validation for Security Vulnerability Repair. AVE Trends in Intelligent Computing Systems. 2026. doi:10.64091/atics.2026.000283

BibTeX
@article{Reddy2026,
  author = {Reddy, L. Yashwanth and Sumanth, M. N. S. and Regin, R. and Bose, S. Rubin and Jeev, S. Sharan and Sherine, S. Benitta},
  title = {SECUREPATCH: Specialized Multi-Agent Architecture with Automated Validation for Security Vulnerability Repair},
  journal = {AVE Trends in Intelligent Computing Systems},
  year = {2026},
  doi = {10.64091/atics.2026.000283}
}
Kostenlos starten →