Query Interface

To query multiple databases using natural language, this Skill leverages LLMs to translate user questions into structured execution plans, validating and running them across document, vector, and graph data stores to synthesize results.

Yes, you can execute a single query plan across document, vector, and graph databases. The system translates natural language into a unified plan, performing document scans, vector searches, and graph traversals seamlessly across multiple stores.

LLM-based query planning for a unified data engine is the process of converting natural language requests into structured, executable plans. It validates these plans and executes them across diverse data stores to generate natural language summaries.

You generate structured insights from natural language data queries by submitting questions to the system. It translates them into query plans, executes them across multiple data stores, and synthesizes the results into natural language summaries and visualizations.

No, this natural language query interface requires no external dependencies. It operates independently to handle LLM integration, query plan validation, multi-store execution, and result synthesis without needing additional setup tools.