@agent-memory/core

Core library for agent-memory — long-term, cross-project memory for AI agents.

Provides memory storage, hybrid search (FTS + vector), embedding, and configuration primitives.

Install

npm install @agent-memory/core

Usage

import { SqliteSource, getDbPath } from '@agent-memory/core';

const source = new SqliteSource(getDbPath());
await source.init();

// Store
const mem = await source.add({
  content: 'Node 22 ESM has path resolution bug in monorepo',
  tags: ['node', 'esm'],
});

// Search (hybrid: FTS + vector)
const results = await source.search('ESM compatibility');

// Retrieve full content
const [full] = await source.get([results[0].id]);

await source.close();

Exports

• SqliteSource — SQLite-backed memory source (FTS5 + sqlite-vec)
• MemorySource — Interface for implementing custom sources
• Types — Memory, AddInput, UpdateInput, SearchOpts, ListOpts, SearchResult, TagCount, SourceStats
• Embedding — EmbeddingProvider, getTransformersProvider() (nomic-embed-text-v1.5, 768 dims, local)
• Search — searchPipeline(), rrfFuse() (Reciprocal Rank Fusion)
• Config — getDbPath(), getConfigDir(), loadConfig(), getConfigValue(), setConfigValue()

Search Pipeline

search(query)
  ├── FTS5 keyword search (always on)
  ├── Vector similarity search (when embeddings available)
  └── RRF Fusion: score = Σ 1/(k + rank_i), k=60
      → merged, deduplicated, sorted
      → post-filter by tags/date

License

MIT