MemoryStack is an AI memory system that provides persistent memory for intelligent agents. It enables AI agents to remember context, learn from interactions, and build knowledge graphs over time.

How does MemoryStack integrate with AI frameworks?

MemoryStack seamlessly integrates with popular AI frameworks including OpenAI, LangChain, CrewAI, LlamaIndex, LangGraph, and more through our Python and Node.js SDKs.

What are the key features of MemoryStack?

Key features include persistent memory storage, adaptive memory consolidation, knowledge graph visualization, semantic search, multi-agent support, RAG integration, and comprehensive API access.

Memories API

Create, retrieve, and manage memories

POST/api/v1/memories

Create new memories from conversation messages. The system automatically extracts semantic facts and stores them with embeddings.

Authentication

Requires API key in Authorization header:

Authorization: Bearer YOUR_API_KEY

Request Body

{
  "messages": [
    {
      "role": "user",
      "content": "I love TypeScript"
    },
    {
      "role": "assistant",
      "content": "TypeScript is great!"
    }
  ],
  "user_id": "user_123",  // Optional
  "metadata": {           // Optional
    "source": "chat",
    "session_id": "sess_xyz"
  }
}

Parameters

Field	Type	Required	Description
messages	array	Yes	Array of conversation messages
user_id	string	No	End user ID for B2B applications
metadata	object	No	Additional context and tags

Response

{
  "success": true,
  "memories_created": 2,
  "memory_ids": [
    "69e0deb6-88d8-49a7-9da7-26a9c8f08d4b",
    "8daf534c-2583-43ee-ba5e-2bbcfee110cb"
  ],
  "owner_id": "user_abc123",
  "user_id": "user_123"
}

Example Request

curl -X POST https://your-domain.com/api/v1/memories \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": "I love TypeScript"
      },
      {
        "role": "assistant",
        "content": "TypeScript is great!"
      }
    ],
    "user_id": "user_123"
  }'

GET/api/v1/memories

List memories with optional filtering and pagination.

Query Parameters

Parameter	Type	Default	Description
user_id	string	-	Filter by user ('self', 'all', or specific ID)
limit	number	20	Max results per page (1-100)
cursor	string	-	Pagination cursor
order	string	desc	Sort order ('asc' or 'desc')
memory_type	string	-	Filter by type (preference, fact, etc.)
min_confidence	number	-	Minimum confidence score (0-1)

Response

{
  "success": true,
  "count": 2,
  "next_cursor": "eyJpZCI6IjEyMyJ9",
  "results": [
    {
      "id": "69e0deb6-88d8-49a7-9da7-26a9c8f08d4b",
      "owner_clerk_id": "user_abc123",
      "end_user_id": "user_123",
      "content": "The user loves TypeScript.",
      "memory_type": "preference",
      "confidence": 0.90,
      "metadata": {
        "source": "chat"
      },
      "source_type": "text",
      "created_at": "2024-11-15T10:30:00Z",
      "updated_at": "2024-11-15T10:30:00Z"
    }
  ]
}

Example Request

curl -X GET "https://your-domain.com/api/v1/memories?user_id=user_123&limit=20" \
  -H "Authorization: Bearer YOUR_API_KEY"

Memory Types

The system automatically classifies memories into types:

preference - User preferences and likes
fact - Factual information
experience - Past experiences and events
goal - Future intentions and goals
relationship - Connections between entities

Error Responses

401 Unauthorized

{
  "error": "Invalid API key"
}

429 Rate Limit Exceeded

{
  "error": "Monthly API limit exceeded"
}

400 Bad Request

{
  "error": "Invalid request payload",
  "details": {
    "fieldErrors": {
      "messages": ["Required"]
    }
  }
}