Skip to main content

Vector Search API

POST /api/v1/memories/search​

Perform semantic search across your memories using vector similarity.

Authentication​

X-API-Key: YOUR_API_KEY

Request Body​

{
  "query": "string (required)",
  "limit": 10,
  "threshold": 0.7,
  "type": "project"
}

Response​

{
  "success": true,
  "data": {
    "results": [
      {
        "id": "mem_abc123",
        "title": "Meeting Notes",
        "content": "Meeting notes from Q4 planning...",
        "memory_type": "meeting",
        "type": "meeting",
        "similarity_score": 0.95,
        "tags": ["planning", "q4"],
        "metadata": {
          "type": "meeting",
          "date": "2024-01-15"
        }
      }
    ],
    "total": 1,
    "query": "Q4 planning decisions",
    "threshold": 0.7
  }
}

Examples​

import { createMemoryClient } from '@lanonasis/memory-client/core';

const client = createMemoryClient({
  apiUrl: 'https://api.lanonasis.com',
  apiKey: process.env.LANONASIS_API_KEY
});

const results = await client.searchMemories({
  query: "Q4 planning decisions",
  limit: 10,
  threshold: 0.8
});
const results = await client.searchMemories({
  query: "budget discussions",
  tags: ["meeting"],
  limit: 5,
  threshold: 0.75
});