Quickstart

Get your AI agent connected to HyperMemory in 5 minutes.

Prerequisites

• A HyperMemory account (sign up free)
• An API key
• An MCP-compatible agent (Claude, OpenClaw, CrewAI, etc.)

Step 1: Get your API key

1. Sign up at app.hypermemory.io
2. Go to Settings → API Keys
3. Click Create New Key
4. Copy your key

Step 2: Connect your agent

Add HyperMemory to your agent’s MCP configuration:

{
  "mcpServers": {
    "hypermemory": {
      "type": "url",
      "url": "https://api.hypermemory.io/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

{
  "mcpServers": {
    "hypermemory": {
      "type": "url",
      "url": "https://api.hypermemory.io/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

from mcp import Client

client = Client(
    server_url="https://api.hypermemory.io/mcp",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

Step 3: Store your first memory

Once connected, your agent can use memory tools:

memory_store(
  content="Our Q3 priority is improving API response times",
  node_type="decision",
  metadata={"quarter": "Q3-2026", "team": "backend"}
)

Response:

{
  "id": "node_abc123",
  "status": "created",
  "created_at": "2026-03-03T10:30:00Z"
}

Step 4: Recall a memory

memory_recall(
  query="What are our Q3 priorities?",
  max_results=5
)

Response:

{
  "results": [
    {
      "id": "node_abc123",
      "content": "Our Q3 priority is improving API response times",
      "node_type": "decision",
      "relevance": 0.95
    }
  ]
}

Step 5: Create a relationship

memory_store(
  content="Sarah Chen is leading the API performance initiative",
  node_type="assignment",
  relationships=["node_abc123"]
)

This creates a new node and automatically links it to the Q3 priority decision.

Step 6: Find related memories

memory_find_related(
  node_id="node_abc123",
  depth=2
)

Response:

{
  "results": [
    {
      "id": "node_def456",
      "content": "Sarah Chen is leading the API performance initiative",
      "node_type": "assignment",
      "distance": 1
    }
  ]
}

Verify everything works

Ask your agent: “What do you remember about Q3 priorities?”

If configured correctly, your agent will use memory_recall to retrieve the stored memory and respond with the information.

What’s next?