Memos Search

Search and retrieve memory content using MemOS API based on user queries, supporting up to top 3 relevant results.

MemOS - External Memory Service

MemOS provides persistent memory storage via REST API.

API Base URL

{{MEMOS_API_URL}}

Available Endpoints

Add Memory

POST /add
Body: {"content": "...", "source": "agent/filename.md"}

Read Memory

GET /read/{agent}/{filename}

List Memories

GET /memories

Search Memories

POST /search
Body: {"query": "...", "top_k": 3}

Delete Memory

DELETE /delete/{agent}/{filename}

List Agents

GET /agents

Health Check

GET /health

Usage Examples

Add memory:

import requests
requests.post('{{MEMOS_API_URL}}/add', json={
    'content': '咖啡大佬今天教我使用MemOS',
    'source': 'alin/2026-02-23.md'
})

Read memory:

import requests
r = requests.get('{{MEMOS_API_URL}}/read/alin/2026-02-23.md')
print(r.json()['content'])

Search memory:

import requests
r = requests.post('{{MEMOS_API_URL}}/search', json={
    'query': '咖啡大佬教了什么',
    'top_k': 3
})
for item in r.json():
    print(f"{item['source']}: {item['score']}")

Delete memory:

import requests
requests.delete('{{MEMOS_API_URL}}/delete/alin/2026-02-23.md')

List all agents:

import requests
r = requests.get('{{MEMOS_API_URL}}/agents')
for agent in r.json()['agents']:
    print(f"{agent['name']}: {agent['file_count']} files")

Notes

• Use Python requests library for API calls
• Always use json= parameter (not raw body) for POST requests
• File path format: {agent}/{filename}.md
• API returns JSON with content field for read operations
• Delete removes both file and vector index
• Run MemOS server: python D:\AI\MemOS\api_server.py