A Framework to Evaluate LLM Agents for Network Configuration

This document specifies an evaluation framework and related definitions for intent-driven network configuration using Large Language Model(LLM)-based agents. The framework combines an emulator-based interactive environment, a suite of representative tasks, and multi-dimensional metrics to assess reasoning quality, command accuracy, and functional correctness. The framework aims to enable reproducible, comprehensive, and fair comparisons among LLM- driven network configuration approaches.