CLI-based RAG (Retrieval-Augmented Generation) system with document search capabilities.
- Interactive Chat Interface: Streaming AI responses powered by Google Gemini
- Document Search: AI can search through uploaded documents to provide informed answers
- Vector Database: Uses libSQL with native vector support for semantic search
Quick start (npm, scoped package):
# Option 1: run the package directly (may run the default bin if supported)
npx @yeoularu/rag-cli
# Option 2: explicitly run the CLI command from the scoped package (recommended)
npx --package=@yeoularu/rag-cli ragUsing Bun:
# Explicitly run the CLI command from the scoped package
bunx @yeoularu/rag-cli ragSpecify a custom DB path (default: ~/.rag-cli/libsql.db):
npx --package=@yeoularu/rag-cli rag --db ~/.rag-cli/mydata.dbLocal development:
bun install
bun run devFormat code:
bun run formatThis project was created using bun init. Bun is a fast all-in-one JavaScript runtime.