VariabLens is an open-source tool that provides an interactive and extensible environment for analyzing feature models (FMs), combining established variability analysis techniques with configurable large language models (LLMs). The tool allows users to upload models expressed in the Universal Variability Language (UVL), including models available from UVLHub, select LLMs for interaction, and explore models through a conversational interface. VariabLens integrates Model Context Protocol (MCP)-based tool execution to ensure that formal operations, such as the computation of metrics including atomic sets and average branching factor, are performed reliably while enabling LLM-driven explanations and summaries. In addition, the tool supports the simultaneous analysis of multiple FMs, facilitating comparison through both structural metrics and semantic insights. To support extensibility, VariabLens incorporates an MCP-based architecture in which a dedicated MCP client coordinates tool execution between LLMs and external analysis services. The platform includes a preconfigured but extensible MCP server exposing UVL-based variability analysis operations, such as metric computation and constraint analysis, through Flamapy-based tools, while enabling the integration of additional domain-specific services and tasks. Although the current implementation focuses on UVL, the architecture is designed to accommodate additional variability modeling formats. By bridging formal FM analysis and AI-driven interaction, VariabLens provides a practical environment for both researchers and practitioners. A video demonstration of the tool is shown in the following link: https://zenodo.org/records/20434200