If you need vector similarity search but don't want to spin up a separate vector database, this gets embeddings into SQLite with minimal fuss. It's a Python wrapper around the sqlite-vec extension, handling the usual pain points like dimension matching and serialization between your embedding model and the database schema. The gates system is smart: it walks you through dimension locks, serialization alignment, and makes you cite docs before trusting any edge case behavior. Supports float32, int8, and bit vectors with KNN queries. At 173 installs it's still small, but the approach is clean for prototypes or apps where SQLite is already your storage layer and adding another service feels like overkill.
npx skills add https://github.com/existential-birds/beagle --skill sqlite-vec