Natural language to SQL for DuckDB

Connect DuckDB and ask your first question

Install the package, register the connection once, build the knowledge base, and query. Leave out --password and it is prompted for without echoing, then stored in your OS keychain rather than the connectors file. Only where no keychain is available does it go into the stored connection URL, with a warning.

$ pip install "nlqueries-core[duckdb]"
$ nlqueries connect duckdb --database /data/warehouse.db --alias duck-local
$ nlqueries process-history duck-local --annotate
$ nlqueries export-kb duck-local
$ nlqueries query duck-local "how many orders were placed last month by first-time customers?"

How NL2SQL works on DuckDB

Most text-to-SQL tools hand the LLM a schema dump and hope. NLQueries builds a YAML knowledge base from your DuckDB schema and from real query history, in this case none; DuckDB is file-based and keeps no query history across connections, so it learns which joins your team actually uses and what your columns mean. That knowledge base is what the model sees when it writes SQL.

Every generated statement is parsed with sqlglot and refused unless it is a single read-only query, and its columns are checked against the schema in the knowledge base. It then runs read-only where the engine allows it, with a statement timeout; what each connector enforces has the details per engine. Use nlqueries ask to see the SQL without running it, or nlqueries query to run it. A semantic cache answers repeated questions without another LLM call, and runs the stored SQL again so the rows are current.

DuckDB-specific details

There is no host, port or password: --database is a file path, and omitting it gives you a transient in-memory database.

Because there is no persisted query history, the knowledge base comes from schema introspection: primary keys via duckdb_constraints(), with foreign keys skipped since analytics workloads rarely declare them. Add relationships to the YAML knowledge base by hand for best results.

A database file is opened read-only, and every connection runs with external access disabled and its configuration locked, so a query cannot read local files through read_csv_auto() and similar table functions. An in-memory database cannot be opened read-only, but the same file-access restrictions apply.

Full connector notes are in the connectors guide, and the read-only role to create is in database hardening. This connector covers DuckDB files and in-memory DuckDB databases.

Use it from Claude, Cursor or any MCP client

Run nlqueries mcp-server start and point Claude Desktop, Cursor, or any Model Context Protocol client at it. Its query tool generates SQL through the same pipeline, runs it read-only and returns the rows. Over stdio, the transport desktop clients use, there is nothing to configure. A network transport will not start without authentication, by OIDC or a pre-shared token, and grants per agent and action decide who may run SQL. See MCP authentication.