Feb 22, 2026

DBMayCry

ai llm sql database documentation

Two-phase system. Database Parsing (top): SQLAlchemy extracts schema from any SQL DB, outputs both JSON and Markdown formats, a Documentation LLM enriches the raw schema with docs, producing a Schema+Docs variant in both formats. This feeds into a TOON (Schema+Docs) representation that becomes the GraphDB input.

Chat Interface (bottom): User queries hit the GraphDB via LangChain's GraphCypherQAChain to retrieve relevant schema context, which alongside chat memory (JSON) gets passed to a Chat LLM that generates the final response.

 
 
 
 
 

This build was uploaded as a hackathon project

Hackathon

HackFest 2.0

View All Projects
Give a star to encourage!Discussion
Start a new conversation!
Login to join the discussion