Use Verba when an operator needs a repeatable RAG chatbot workspace for exploring a specific dataset before summarization, support-answer drafting, retrieval QA, or agent context handoff. The workflow is to install or deploy Verba, configure the Weaviate backing store and model/provider keys, ingest an explicit corpus, test representative questions, inspect retrieved context and answer quality, then decide whether the dataset is ready for downstream agent use. Invoke this instead of using a generic hosted chatbot when the work needs a local or controlled RAG environment with visible ingestion, retrieval settings, source documents, and reviewable answers. Keep the corpus, provider choices, and test questions recorded so later reviewers can reproduce the run. The scope boundary is document ingestion, retrieval configuration, and answer review for a named corpus. It is not a generic Weaviate listing, an autonomous agent framework, or an unsupervised ingestion pipeline for arbitrary private documents.