This is a quick start guide.For full documentation and advanced features:docs.memof.ai

QUICK START GUIDE

Get started with Memory-of-Agents in minutes. Essential API calls and examples to build intelligent applications with persistent memory.

ESSENTIAL API CALLS

AUTHENTICATION

Include your API key in the Authorization header for all requests:

# Set your API key
export MEMOF_API_KEY="..."

# All requests use this header
curl -H "Authorization: Bearer $MEMOF_API_KEY" \
     -H "Content-Type: application/json" \
     https://api.beta.memof.ai/v1/memories

STORE MEMORY

Store any information in the dual-track memory system:

curl -X POST "https://api.beta.memof.ai/v1/memories" \
  -H "Authorization: Bearer $MEMOF_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "User John prefers dark mode and uses keyboard shortcuts",
    "metadata": {
      "user_id": "john_doe",
      "type": "preference",
      "priority": "high",
      "tags": ["ui", "accessibility", "shortcuts"]
    },
    "context": {
      "session_id": "session_abc123",
      "conversation_id": "conv_456",
      "timestamp": "2024-08-02T10:30:00Z"
    }
  }'

# Response
{
  "id": "mem_7x9k2m",
  "status": "stored",
  "tracks": {
    "raw": "verified",
    "semantic": "processed"
  }
}

SEARCH MEMORIES

Retrieve relevant memories using hybrid search:

# Search with natural language
curl -X GET "https://api.beta.memof.ai/v1/memories/search" \
  -H "Authorization: Bearer $MEMOF_API_KEY" \
  -G -d "query=user interface preferences" \
  -d "user_id=john_doe" \
  -d "limit=10"

# Advanced search with weights
curl -X GET "https://api.beta.memof.ai/v1/memories/search" \
  -H "Authorization: Bearer $MEMOF_API_KEY" \
  -G -d "query=dark mode settings" \
  -d "vector_weight=0.7" \
  -d "keyword_weight=0.3" \
  -d "tags=ui,accessibility"

# Response
{
  "memories": [
    {
      "id": "mem_7x9k2m",
      "content": "User John prefers dark mode...",
      "relevance_score": 0.94,
      "metadata": {...},
      "created_at": "2024-08-02T10:30:00Z"
    }
  ],
  "total": 1,
  "search_time_ms": 45
}

UPDATE MEMORY

Update existing memories while preserving history:

curl -X PUT "https://api.beta.memof.ai/v1/memories/mem_7x9k2m" \
  -H "Authorization: Bearer $MEMOF_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "User John prefers dark mode, uses vim shortcuts, and likes compact layouts",
    "metadata": {
      "tags": ["ui", "accessibility", "shortcuts", "layout"],
      "priority": "high"
    }
  }'

COMMON USE CASES

CHATBOT MEMORY

Give your AI assistant perfect memory of user preferences and conversation history.

# Store user preference
POST /memories → "User likes concise responses"
# Later retrieve context
GET /search → "communication style"

PERSONALIZATION

Build apps that remember and adapt to individual user behaviors and preferences.

# Track user behavior
POST /memories → "Clicked 'advanced mode'"
# Personalize experience
GET /search → "interface preferences"

ANALYTICS & INSIGHTS

Store and analyze user interactions to discover patterns and improve your product.

# Store interaction data
POST /memories → "Feature X used 5 times"
# Analyze patterns
GET /search → "feature usage patterns"

KNOWLEDGE BASE

Create intelligent knowledge systems that learn and remember information over time.

# Store knowledge
POST /memories → "API rate limit is 1000/hour"
# Retrieve when needed
GET /search → "rate limits"

READY TO BUILD?

Questions? Email us at developers@memof.ai