langchain-schemagate

A LangChain retriever for text-to-SQL. Each question gets only the tables it needs, picked from the tables the caller may read. A table the caller can't read is never a candidate, so it can't reach the prompt.

pip install langchain-schemagate

from schemagate import Catalog, Principal
from langchain_schemagate import SchemagateRetriever, prompt_fragment

cat = Catalog().bootstrap("postgresql://localhost/app")
retriever = SchemagateRetriever(catalog=cat, top_k=6,
                                principal=Principal("okta:jdoe", roles={"finance"}))

docs = retriever.invoke("revenue by month")      # one Document of DDL per selected table
schema_for_prompt = prompt_fragment(docs)

It is a BaseRetriever, so it composes: retriever | RunnableLambda(prompt_fragment) | prompt | llm. Selection runs offline (BM25 plus a small embedder) and calls no model.

On Spider 2.0's public BigQuery and Snowflake schemas, the median database with 50+ tables went from the full schema to a selection 91.1% smaller, in prompt tokens. On small databases there is little to cut: 8.8% on BIRD's seven-table ones. The script and the recall trade-off are in schemagate's BENCHMARKS.md.

The retriever itself is maintained in schemagate, which also has the full guide: docs/langchain.md. This package is the langchain-<name> distribution LangChain lists integrations under, and it runs LangChain's standard retriever tests.

Apache-2.0.