SPARQL has long been the standard interface to knowledge graphs—but its technical complexity limits access for many audiences. This webinar introduced Quagga, an LLM-based agent developed within GRAPHIA that enables conversational exploration of knowledge graphs using natural language. We presented how the agent works, including methods to automatically generate metadata from any knowledge graph to support text-to-SPARQL translation. The session also highlighted the Quagga benchmark, an ongoing collaborative crowdsourcing effort to build a knowledge graph question-answering dataset focused on the social sciences and humanities. A recording of this webinar is avaliable here. Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the Agency. Neither the European Union nor the granting authority can be held responsible for them.