reasonkit-mem 0.1.7

High-performance vector database & RAG memory layer - hybrid search, embeddings, RAPTOR trees, BM25 fusion, and semantic retrieval for AI systems
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

//! Docset ingestion (Cursor-like `@Docs`) for ReasonKit Mem.
//!
//! A **docset** is a named documentation source identified by a start URL and
//! one or more allowed URL prefixes. `reasonkit-mem` can crawl docsets,
//! normalize pages to Markdown, chunk them, and index them for retrieval.
//!
//! This module is **self-contained**: refresh scheduling works without
//! `reasonkit-org` by running `rk-mem docs refresh --due` periodically
//! (e.g., via systemd timer/cron) or by embedding the scheduler in your own
//! process.

mod ingest;
mod mcp_server;
mod store;

pub use store::{Docset, DocsetId, DocsetStore, DocsetStoreConfig, RefreshPolicy, RefreshStatus};

pub use ingest::{
    open_default_docset_retriever, DocsetIngestOptions, DocsetIngestReport, DocsetIngestor,
};

pub use mcp_server::DocsetMcpServer;