I prompt, therefore i program: An agent layer for translating prompts into calls to the 5G core operations API
Abstract
5G networks have introduced advanced programmability mechanisms through standardized APIs, most notably the Network Exposure Function (NEF), which enables the controlled exposure of core network capabilities to external applications. However, the semantic and structural complexity of these APIs still represents a significant barrier to their adoption and automation. In this paper, we investigate the use of AI Agents to perform calls to the NEF Traffic Influence API in the 5G core, evaluating the ability of Large Language Models (LLMs) to translate high-level requests into correct interactions with a 5G system component. The proposed solution leverages the Model Context Protocol (MCP) to integrate the models with tools that encapsulate the NEF Traffic Influence API. We evaluate different LLMs, including Qwen, GPT-OSS-20B, and Claude Sonnet 4.5, considering variations in model size and different prompt engineering strategies. The evaluations were conducted in an environment based on free5GC and UERANSIM, using metrics such as success rate, token consumption, and response time.
References
3GPP (2024). Management and orchestration; Intent driven management services for mobile networks. Technical Report TS 28.312, 3GPP.
Bimo, F. A., Galdon, M. A. C., Lai, C.-K., Cheng, R.-G., and Chong, E. K. (2025). IntentBased Network for RAN Management with Large Language Models. arXiv preprint arXiv:2507.14230.
Brodimas, D., Birbas, A., Kapolos, D., and Denazis, S. (2025). Intent-Based Infrastructure and Service Orchestration Using Agentic-AI. IEEE Open Journal of the Communications Society, 6:7150–7168.
Brodimas, D., Trantzas, K., Agko, B., Tziavas, G. C., Tranoris, C., Denazis, S., and Birbas, A. (2024). Towards Intent-based Network Management for the 6G System adopting Multimodal Generative AI. In 2024 Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit), pages 848–853. IEEE.
GSMA (2023). CAMARA Project: Network APIs for Developers. [link].
Habib, M. A., Iturria-Rivera, P. E., Ozcan, Y., Elsayed, M., Bavand, M., Gaigalas, R., and Erol-Kantarci, M. (2025). Harnessing the Power of LLMs, Informers and Decision Transformers for Intent-driven RAN Management in 6G. IEEE Transactions on Network Science and Engineering.
Hou, X., Zhao, Y., Wang, S., and Wang, H. (2025). Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions. arXiv preprint arXiv:2503.23278.
Khan, Z., Hussain, A., Thakur, M., Hellas, A., and Papadimitratos, P. (2025). NEFMind: Parameter-Efficient Fine-Tuning of Open-Source LLMs for Telecom APIs Automation.
Ksentini, A. and Frangoudis, P. A. (2020). Toward slicing-enabled multi-access edge computing in 5g. IEEE Network, 34(2):99–105.
Manias, D. M., Chouman, A., and Shami, A. (2024). Towards Intent-Based Network Management: Large Language Models for Intent Extraction in 5G Core Networks. In 2024 20th International Conference on the Design of Reliable Communication Networks (DRCN), pages 1–6. IEEE.
Peterson, L. and Sunay, O. (2020). 5G Mobile Networks: A Systems Approach. Morgan & Claypool Press, 1 edition.
Sapkota, R., Roumeliotis, K. I., and Karkee, M. (2026). Ai agents vs. agentic ai: A conceptual taxonomy, applications and challenges. Information Fusion, 126:103599.
The Linux Foundation and CAMARA Project (2025). In Concert: Bridging AI Systems Network Infrastructure through MCP: How to Build Network-Aware Intelligent Applications. Technical report. 12 pages.
Zhou, H., Hu, C., Yuan, Y., Cui, Y., Jin, Y., Chen, C., Wu, H., Yuan, D., Jiang, L., Wu, D., Liu, X., Zhang, J., Wang, X., and Liu, J. (2025). Large Language Model (LLM) for Telecommunications: A Comprehensive Survey on Principles, Key Techniques, and Opportunities. IEEE Communications Surveys & Tutorials, 27(3):1955–2005.
