slm-remember

Save coding decisions locally for semantic search by storing content in an SQLite database with automatic TF-IDF indexing, ensuring fast retrieval across sessions. The system links related entities using a knowledge graph for structured context retrieval.

A knowledge graph connects related memories and entities in local storage by automatically linking stored content during indexing. This structured approach allows fast semantic search across personal notes and coding decisions using TF-IDF vectors.

Yes, you need SuperLocalMemory v3 installed to use local knowledge graph storage. The system uses this prerequisite environment to manage the SQLite database and execute TF-IDF vector searches for remembering content.

Yes, you can use SQLite and TF-IDF vectors to remember project context across different tools. The local storage architecture supports cross-tool workflows by indexing notes and linking entities for fast semantic retrieval.

The best way to index personal notes with a local knowledge graph is using automatic indexing with TF-IDF vectors and SQLite storage. This method connects related entities automatically, enabling fast semantic search for future retrieval.

Local semantic search requires TF-IDF vectors and a knowledge graph to enable fast structured retrieval. TF-IDF vectors handle the semantic search scoring in SQLite, while the knowledge graph links related entities to organize the stored context effectively.