@mastra/pg

SQLite implementation for Mastra, providing both vector similarity search and general storage capabilities with connection pooling and transaction support.

Installation

npm install @mastra/libsql

Usage

Vector Store

import { LibSQLVector } from '@mastra/libsql';

const vectorStore = new LibSQLVector({
  url: 'file:./my-db.db'
});

// Create a new table with vector support
await vectorStore.createIndex({
  indexName: 'my_vectors',
  dimension: 1536,
  metric: 'cosine',
});

// Add vectors
const ids = await vectorStore.upsert({
  indexName: 'my_vectors',
  vectors: [[0.1, 0.2, ...], [0.3, 0.4, ...]],
  metadata: [{ text: 'doc1' }, { text: 'doc2' }],
});

// Query vectors
const results = await vectorStore.query({
  indexName: 'my_vectors',
  queryVector: [0.1, 0.2, ...],
  topK: 10, // topK
  filter: { text: 'doc1' }, // filter
  includeVector: false, // includeVector
  minScore: 0.5, // minScore
});

Storage

import { LibSQLStore } from '@mastra/pg';

const store = new LibSQLStore({
  url: 'file:./my-db.db',
});

// Create a thread
await store.saveThread({
  id: 'thread-123',
  resourceId: 'resource-456',
  title: 'My Thread',
  metadata: { key: 'value' },
});

// Add messages to thread
await store.saveMessages([
  {
    id: 'msg-789',
    threadId: 'thread-123',
    role: 'user',
    type: 'text',
    content: [{ type: 'text', text: 'Hello' }],
  },
]);

// Query threads and messages
const savedThread = await store.getThread('thread-123');
const messages = await store.getMessages('thread-123');

Configuration

The LibSQLStore store can be initialized with:

• Configuration object with url and auth. Auth is only necessary when using a provider like Turso

Features

Vector Store Features

• Vector similarity search with cosine, euclidean, and dot product metrics
• Advanced metadata filtering with MongoDB-like query syntax
• Minimum score threshold for queries
• Automatic UUID generation for vectors
• Table management (create, list, describe, delete, truncate)

Storage Features

• Thread and message storage with JSON support
• Atomic transactions for data consistency
• Efficient batch operations
• Rich metadata support
• Timestamp tracking
• Cascading deletes

Supported Filter Operators

The following filter operators are supported for metadata queries:

• Comparison: $eq, $ne, $gt, $gte, $lt, $lte
• Logical: $and, $or
• Array: $in, $nin
• Text: $regex, $like

Example filter:

{
  $and: [{ age: { $gt: 25 } }, { tags: { $in: ['tag1', 'tag2'] } }];
}

Vector Store Methods

• createIndex({indexName, dimension, metric?, indexConfig?, defineIndex?}): Create a new table with vector support
• upsert({indexName, vectors, metadata?, ids?}): Add or update vectors
• query({indexName, queryVector, topK?, filter?, includeVector?, minScore?}): Search for similar vectors
• defineIndex({indexName, metric?, indexConfig?}): Define an index
• listIndexes(): List all vector-enabled tables
• describeIndex(indexName): Get table statistics
• deleteIndex(indexName): Delete a table
• truncateIndex(indexName): Remove all data from a table

Storage Methods

• saveThread(thread): Create or update a thread
• getThread(threadId): Get a thread by ID
• deleteThread(threadId): Delete a thread and its messages
• saveMessages(messages): Save multiple messages in a transaction
• getMessages(threadId): Get all messages for a thread
• deleteMessages(messageIds): Delete specific messages

Related Links

• LibSQL Documentation