QueryWeaver

Overview

QueryWeaver is an open‑source Text2SQL server that transforms natural‑language questions into executable SQL statements by leveraging a graph‑powered understanding of database schemas. It addresses the perennial challenge developers face when working with relational data: bridging the gap between human intent and the rigid syntax of SQL. By presenting a simple, REST‑based interface (and optional MCP surface), QueryWeaver lets teams ask questions in plain English and instantly receive both the generated query and its results, eliminating the need for manual query construction or extensive database knowledge.

The core value proposition lies in its schema‑aware generation. Instead of treating the database as a black box, QueryWeaver builds an internal graph that captures tables, columns, relationships, and constraints. This context allows the model to produce accurate joins, filter conditions, and aggregation logic that respect referential integrity and cardinality. For developers, this means fewer runtime errors, more reliable data retrieval, and a smoother integration into existing tooling such as BI dashboards or conversational agents.

Key capabilities include:

• Database discovery – and operations let an AI assistant enumerate available data sources and establish connections on demand.
• Schema introspection – returns a rich, machine‑readable description of tables and relationships, enabling downstream tools to build autocomplete or validation logic.
• Query generation & execution – accepts a natural‑language prompt and returns the generated SQL along with result rows, supporting iterative refinement if needed.
• MCP compatibility – The server can expose these operations as MCP endpoints or act as an MCP client, making it easy to plug into larger AI workflows that rely on a shared context layer.

Typical use cases span from chat‑based data exploration (e.g., a customer support bot answering “How many orders did we receive last month?”) to automated reporting (generating SQL for scheduled dashboards) and development tooling (providing instant query suggestions within IDEs). By abstracting away the intricacies of schema design, QueryWeaver empowers both data scientists and software engineers to focus on business logic rather than query syntax.

In practice, a developer can spin up QueryWeaver via Docker or integrate it into an existing application stack. Once running, an AI assistant can call the MCP endpoints to discover databases, fetch schemas, and request queries—all within a single conversational turn. The resulting SQL is guaranteed to be syntactically correct and semantically aligned with the underlying graph, reducing debugging cycles and accelerating time‑to‑value for data‑driven projects.