Skip to main content

Crate memori_core

Crate memori_core 

Source
Expand description

Persistent memory for AI coding agents — SQLite + FTS5 + vector search in a single binary.

memori-ai-core is the Rust engine behind the py-memori Python package and the memori CLI. It stores memories (text + 384-dim embeddings + JSON metadata + access stats) in one SQLite file with WAL journaling, exposes FTS5 full-text search and brute-force cosine vector search, and fuses them with Reciprocal Rank Fusion (k=60).

See https://github.com/archit15singh/memori for the full design and CLI usage.

Re-exports§

pub use types::InsertResult;
pub use types::Memory;
pub use types::MemoriError;
pub use types::Result;
pub use types::SearchQuery;
pub use types::SortField;

Modules§

embed
schema
search
storage
types
util

Structs§

Memori