A tiny local RAG baseline using only SQLite search

JAS RAG is a small local tool for testing before adding heavier parts. It indexes Markdown files, splits them into documents and chunks, and stores everything in SQLite. It uses SQLite FTS5 for , searches at the chunk level, and formats the matched chunks as context for a .

It does not need embeddings, a , rerankers, s, or API keys. The workflow is to index a small folder, run a few questions, check whether the top 1 and top 3 results are right, and inspect the returned chunks directly. Vector search is added only if the test shows SQLite FTS5 is not good enough.

It is not a full RAG platform or production framework, but a simple local baseline for checking .

Key points

  • Markdown files are stored locally in SQLite and searched with SQLite FTS5.
  • The tool avoids embeddings, a , external services, and API keys.
  • Retrieved chunks are formatted into context for a .
  • Top 1 and top 3 search results are checked before adding more complex search.
  • The goal is a simple RAG baseline, not a platform.
Read original