A LangGraph SQL agent that only sees the tables it needs

A new LangGraph example in the open-source schemagate project cuts schema tokens sent to the model by 68% to 78% on a bundled 42-table demo database, sending 489 to 706 tokens per question instead of 2,199, with all three test answers correct using Claude Haiku. The four-node agent uses a SchemagateRetriever to pick only the tables a question needs in an offline step with no model call, then writes one SELECT, runs it, and returns a one-sentence answer; on databases with 50 or more tables the saving reaches about 91%. Schemagate, which is Apache-2.0 and listed in LangChain's official retriever integrations, ranks #1 on the BIRD Mini-Dev leaderboard across SQLite, PostgreSQL and MySQL, and the example installs via pip install langchain-schemagate langgraph langchain-anthropic tiktoken.

A LangGraph SQL agent that only sees the tables it needs Most text-to-SQL agents paste the whole schema into every prompt. On a real database, most of it is tables the question never touches, and you pay for them on every call. I added a small LangGraph example to schemagate that does it differently. It has four nodes: select schema: SchemagateRetriever p icks the few tables the question needs. This step runs offline, with no model call. write sql: the model writes one SELECT using only those tables. run sql: the agent runs the query. If it fails, the error goes back to the model once. answer: the model answers in one sentence from the returned rows. On the bundled 42-table demo database: Question Schema tokens sent Which customers owe us money? 489 vs 2,199 78% fewer How many orders did each carrier ship? 706 vs 2,199 68% fewer Which products are below their reorder level, and in which warehouse? 563 vs 2,199 74% fewer All three answers were correct with Claude Haiku. On databases with 50+ tables the saving reaches about 91%. The selection also respects roles: a table the caller can't read never appears in the prompt. pip install langchain-schemagate langgraph langchain-anthropic tiktoken python examples/langgraph sql agent.py To use your own database, swap demo catalog for Catalog .bootstrap "postgresql://..." . schemagate is 1 on the BIRD Mini-Dev leaderboard SQLite, PostgreSQL, MySQL , is listed in LangChain's official retriever integrations, and is Apache-2.0. Example: examples/langgraph sql agent.py Key Takeaways - •This step runs offline, with no model call. - •This story was reported by Dev.to , covering developments in the dev space. - •AI advancements continue to reshape industries — read the full article on Dev.to for complete coverage. 📖 Continue reading the full article: Read Full Article on Dev.to → https://dev.to/ashish sinha 5241c7673d93/a-langgraph-sql-agent-that-only-sees-the-tables-it-needs-23no