Orclever Proceedings of Research and Development

A Modular Semantic Kernel Agent for Automated Code Review and Refactoring Feedback

Semih Yazıcı

Boyner

https://orcid.org/0009-0006-4503-0256

Seza Dursun

Boyner

https://orcid.org/0000-0003-1389-072X

Bahar Önel

Boyner

https://orcid.org/0009-0007-4597-6591

Tülin Işıkkent

Boyner

https://orcid.org/0009-0005-5775-0093

Sedat Çelik

Boyner

https://orcid.org/0009-0003-0335-6440

Erem Karalar

Boyner

https://orcid.org/0000-0001-6289-9275

Mert Alacan

Boyner

https://orcid.org/0000-0003-3893-6309

DOI: https://doi.org/10.56038/oprd.v7i1.739

Keywords: Code Review, Semantic Kernel, Plugin Orchestration, Refactoring, Large Language Models, Agentic AI, Retrieval-Augmented Generation, Prompt Engineering

---

Abstract

In modern software development, maintaining clean, efficient, and reliable code is critical to team productivity and product quality. This paper introduces a modular Large Language Model (LLM)-based agent, designed using Microsoft’s Semantic Kernel framework, for automated code review and refactoring feedback. The agent leverages plugin-based function orchestration, Retrieval-Augmented Generation (RAG), and dynamic prompt engineering to analyze source code across multiple dimensions; including readability, efficiency, security, and adherence to best practices. Integrated into CI/CD pipelines and broader SDLC workflows, the system provides contextual insights, the system provides contextual insights, suggests specific improvements, and explains reasoning for each recommendation. Evaluation results across real-world open-source repositories demonstrate the agent’s effectiveness in reducing human review time while improving refactor quality. The modular design ensures adaptability to various programming languages and enterprise development environments. This research highlights the potential of agentic LLM systems to augment software engineering workflows with intelligent, transparent, and developer-aligned feedback mechanisms.

Keywords:   Code Review, Semantic Kernel, Plugin Orchestration, Refactoring, Large Language Models, Agentic AI, Retrieval-Augmented Generation, Prompt Engineering

---

References

Microsoft. Semantic Kernel Documentation. https://learn.microsoft.com/en-us/semantic-kernel

OpenAI. Function Calling and Tool Use. https://platform.openai.com/docs/guides/function-calling

LangChain. Agents and Tool Use in LLM Applications. https://docs.langchain.com/docs/components/agents/

GitHub Copilot. “Your AI Pair Programmer.” https://github.com/features/copilot

Raza, M., & Rasool, G. (2023). "LLM-Guided Software Development: Opportunities and Threats." IEEE Software, 40(3), 27–35. https://doi.org/10.1109/MS.2023.3246537

Jain, A., et al. (2022). "CodeBERT: A Pre-Trained Model for Programming and Natural Languages." EMNLP Findings.

Li, X., et al. (2023). "RefactorGPT: Code Refactoring via LLM Agents." arXiv preprint arXiv:2307.09771.

Fernandes, P., et al. (2023). "Improving Code Review Quality with AI Agents: A Modular Pipeline." ACM Transactions on Software Engineering and Methodology.

Kim, H., et al. (2023). "AI-Augmented Refactoring for Legacy Systems: An Empirical Study." SoftwareX, 21, 101322. DOI: https://doi.org/10.1016/j.softx.2023.101322

Wang, Z., et al. (2023). "Human-AI Co-Pilot Systems in Software Engineering: Design Principles and Case Studies." ICSE '23: Proceedings of the 45th International Conference on Software Engineering, 185–196.

Chen, M., et al., “Evaluating Large Language Models Trained on Code,” arXiv:2107.03374, 2021