Irene Denazi

Christos Tranoris, Besiana Agko, Kostis Trantzas et al.

This work proposes an agentic, intent-driven end-to-end (E2E) orchestration framework that integrates intent co-creation with a Test-Driven Quality Assurance paradigm. In this framework, autonomous agents iteratively refine a user's initial intent into a confirmed, auditable specification. Furthermore, the system automatically derives validation tests from these intents before provisioning, directly mirroring the Test-Driven Development workflow in software engineering to ensure proactive Service Level Agreement (SLA) compliance. The architecture is grounded in a standards-aligned knowledge representation using TM Forum (TMF) information models and catalogs. This enables deterministic graph traversal from high-level Product Offerings down to granular Service/Resource and Test specifications. We prototyped this architecture by extending OpenSlice with a message-driven, multi-agent pattern and integrating MCP-enabled (Model Context Protocol) tool access for real-time knowledge retrieval. Currently, our evaluation of the agents targets the intent co-creation phase as a baseline toward full-scale orchestration. Building on experiments with multiple open-source Large Language Model (LLM) backends integrated with the TMF-based knowledge base, we observe substantial variability in tool-use reliability and hallucination patterns, underscoring the critical importance of robust knowledge integration in agentic 6G systems.

Kostis Trantzas, Besiana Agko, Christos Tranoris et al.

6G network complexity necessitates high levels of autonomy, yet current intent-based systems struggle with ambiguous or incomplete human requests. This paper introduces an agent-based, intent-driven end-to-end (E2E) orchestration framework designed for Network-as-a-Service (NaaS) delivery through collaborative intent co-creation. The proposed system leverages a pool of Domain Expert Agents and a TM Forum-aligned Body-of-Knowledge (BoK) to iteratively refine user requests into deterministic, machine-readable actions. A fundamental design principle is the decoupling of cognition and actuation, where AI-driven reasoning is isolated from standardized execution controllers to ensure safety and operational trust. The framework includes a dual-layer memory system to maintain coherence during multi-step collaborations. The presented prototype, built on ETSI OpenSlice and the Model Context Protocol (MCP), evaluates across several open-source Large Language Models (LLMs). While these models demonstrate high instruction compliance, results reveal a significant gap in translating high-resolution intents into valid, catalog-backed orders without hallucinations.