Policy-Governed Agentic Autonomous Quality Assurance for Enterprise Systems
DOI:
https://doi.org/10.22399/ijcesen.4933Keywords:
Agentic AI, Policy-as-Code (PaC), SEPGA Architecture, Open Policy Agent (OPA), Autonomous Software Testing, Agentic DebtAbstract
A paradigm shift of deterministic, script-based automation by probabilistic, agentic autonomous systems characterizes the enterprise software landscape of 2025. This research paper offers a comprehensive analysis of Policy-Governed Agentic Autonomous Quality Assurance (QA), a new architectural design that is aimed at alleviating the stochastic risks that any Large Language Model (LLM) agent presents and exploiting their reasoning abilities. Although traditional automation frameworks are said to have reached a scalability limit due to maintenance overheads that consume over 80 percent of QA resources, agentic systems that are not guarded are said to introduce agentic debt a form of technical debt manifested by autonomous decisions that are not aligned. By performing a strict study of the SEPGA (Self-Evolving, Policy-Governed Agentic Automation) architecture, they show that the incorporation of Policy-as-Code (PaC) engines forms a "policy-bounded autonomy." The quantitative evidence shows that by using policy-governed architectures, compliance violations are minimized by 68 percent, and the cost of maintaining the tests is reduced by about 90 percent. It is concluded in the paper that the future of enterprise reliability is in the cryptographic verification of the layers of governance which bind autonomous agents.
References
1. Bouzenia, I., Devanbu, P., & Pradel, M. (2024). RepairAgent: An autonomous, LLM-based agent for program repair. arXiv. https://doi.org/10.48550/arXiv.2403.17134
2. Chen, W., Su, Y., Zuo, J., Yang, C., Yuan, C., Chan, C.-M., Yu, H., Lu, Y., Hung, Y.-H., Qian, C., Qin, Y., Cong, X., Xie, R., Liu, Z., Sun, M., & Zhou, J. (2024). AgentVerse: Facilitating multi-agent collaboration and exploring emergent behaviors. In Proceedings of the Twelfth International Conference on Learning Representations (ICLR). https://doi.org/10.48550/arXiv.2308.10848
3. Hong, S., Zhuge, M., Chen, J., Zheng, X., Cheng, Y., Zhang, C., Wang, J., Wang, Z., Yau, S. K. S., Lin, Z., Zhou, L., Ran, C., Xiao, L., Wu, C., & Schmidhuber, J. (2024). MetaGPT: Meta programming for a multi-agent collaborative framework. In Proceedings of the Twelfth International Conference on Learning Representations (ICLR). https://doi.org/10.48550/arXiv.2308.00352
4. Hu, S., Zhu, F., Chang, X., & Liang, X. (2021). UPDeT: Universal multi-agent reinforcement learning via policy decoupling with transformers. arXiv. https://doi.org/10.48550/arXiv.2101.08001
5. Hu, V. C., Ferraiolo, D., Kuhn, R., Schnitzer, A., Sandlin, K., Miller, R., & Scarfone, K. (2014). Guide to attribute-based access control (ABAC): Definition and considerations. NIST Special Publication (SP 800-162). National Institute of Standards and Technology.
6. Jackson, F. (2025). AI-native software engineering and vibe coding: A strategic technical survey for modern engineering leadership (2025–2035). SSRN Working Paper.
7. Jeng, J. J., Schiefer, J., & Chang, H. (2003). An agent-based architecture for analyzing business processes of real-time enterprises. In Proceedings of the IEEE International Enterprise Distributed Object Computing Conference (pp. 381–390).
8. Jimenez, C. E., Yang, J., Wettig, A., Yao, S., Pei, K., Press, O., & Narasimhan, K. (2024). SWE-bench: Can language models resolve real-world GitHub issues? In Proceedings of the Twelfth International Conference on Learning Representations (ICLR). https://doi.org/10.48550/arXiv.2310.06770
9. Liu, Y., Zhang, R., Luo, H., Lin, Y., Sun, G., & Niyato, D. (2025). Secure multi-LLM agentic AI and agentification for edge general intelligence by zero trust: A survey. arXiv preprint.
10. Lv, W., Xia, X., & Huang, S. J. (2024). CodeACT: Code adaptive compute-efficient tuning framework for code LLMs. arXiv. https://doi.org/10.48550/arXiv.2408.02193
11. Parasa, M. (n.d.). Policy-centric AI control architectures for enterprise software platforms: A governance framework for SAP systems. Unpublished manuscript.
12. Qian, C., Liu, W., Liu, H., Chen, N., Dang, Y., Li, J., Yang, C., Chen, W., Su, Y., Cong, X., Xu, J., Li, D., Liu, Z., & Sun, M. (2024). ChatDev: Communicative agents for software development. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL) (pp. 15061–15087). https://doi.org/10.18653/v1/2024.acl-long.810
13. Shah, J. K. (2025). Agentic AI for autonomous micro-frontend user interfaces and microservices evolution in cloud platforms. Journal of Computer Science and Technology.
14. Uzair, M. Y., Rehan, M., & Chand, M. (2025). Policy-governed self-evolving architecture for autonomous AI agents in enterprise systems. International Journal of Advanced Research in Computer Science and Software Engineering.
15. Vongalis-Macrow, A. (2007). I, teacher: Re-territorialization of teachers’ multi-faceted agency in globalized education. British Journal of Sociology of Education, 28(4), 425–439.
16. Wang, X., Chen, Y., Yuan, L., Zhang, Y., Li, Y., Peng, H., & Ji, H. (2024). Executable code actions elicit better LLM agents. arXiv. https://doi.org/10.48550/arXiv.2402.01030
17. Zhang, K., Antoniades, A., Örwall, A., Xie, Y., Goyal, A., & Wang, W. Y. (2024). SWE-search: Enhancing software agents with Monte Carlo tree search and iterative refinement. arXiv. https://doi.org/10.48550/arXiv.2410.20285
18. Zhang, Y., Ruan, H., Fan, Z., & Roychoudhury, A. (2024). AutoCodeRover: Autonomous program improvement. In Proceedings of the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA) (pp. 1–12). https://doi.org/10.48550/arXiv.2404.05427
19. Zwerdling, N., Boaz, D., & Rabinovich, E. (2025). Towards enforcing company policy adherence in agentic workflows. In Proceedings of the International Conference on Computational Linguistics and Intelligent Text Processing.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
