Guide
Python SDK
JS SDK
HTTP API
MCP Server
Guide
Python SDK
JS SDK
HTTP API
MCP Server

Memories

Store and search information using semantic understanding.

Store Memory

POST /api/integrations/memory/store

Request:

{
  "bot_id": "bot_123",
  "content": "User prefers email. Premium since 2023.",
  "metadata": {"category": "preference", "user_id": "user_456"}
}

Example:

curl -X POST https://api.memof.ai/api/integrations/memory/store   -H "Authorization: Bearer moa_your_token"   -H "Content-Type: application/json"   -d '{
    "bot_id": "bot_123",
    "content": "User prefers dark mode and TypeScript",
    "metadata": {"category": "preference"}
  }'

List Memories

GET /api/integrations/memory/{bot_id}/list?page=1&per_page=20

Search Memories

Semantic search using natural language.

POST /api/integrations/memory/search

Request:

{
  "bot_id": "bot_123",
  "query": "communication preferences",
  "limit": 5,
  "metadata_filter": {"category": "preference"}
}

Example:

curl -X POST https://api.memof.ai/api/integrations/memory/search   -H "Authorization: Bearer moa_your_token"   -H "Content-Type: application/json"   -d '{
    "bot_id": "bot_123",
    "query": "users who prefer phone calls",
    "limit": 10
  }'

Delete Memory

DELETE /api/integrations/memory/{memory_id}/delete

Best Practices

✅ Good Content:

{
  "content": "Sarah (sarah@company.com) is CTO at TechCorp. Interested in enterprise API."
}

❌ Poor Content:

{
  "content": "User likes stuff"
}

Search Strategies

Natural Language:

  • "customers who prefer email"
  • "technical issues this week"
  • "feature requests from enterprise users"

With Filters:

{
  "query": "support tickets",
  "metadata_filter": {"priority": "high", "status": "open"}
}

Next Steps

ON THIS PAGE