Academy
⌘K
Getting Started
  • Platform Overview
  • AWS Marketplace Setup
  • Quickstart Guide
Connections
  • Overview
Workflow Builder
  • Overview
  • Nodes Catalog
Helix AI Assistant
  • Overview
Mapping Engine
  • Overview
Workflow Library
  • Overview
  • HL7
  • FHIR
  • Insurance
  • Agentic
  • Mapping
Security & Infrastructure
  • Architecture
  • HIPAA Compliance
  • Performance & Scalability
← Back to ConnectHealth
Get Started
Back to Nodes
AI Agent1 input → 1 output

AI Memory (In-Mem)

payload (object |…)topic (string)AI AgentAI Memory (In-Mem)payload (object)aimemory (object)Data flowProperty

Stores AI conversation memories in volatile memory. Fast but data is lost on Node-RED restart.

Inputs

payload object | string
Data to store or search query
topic string
Memory operation: "store", "search", "count", or empty for get-all

Outputs

payload object
Operation result (varies by topic)
aimemory object
Memory context: {context: [...], maxItems}

Configuration

Max Items number
Maximum memories to store. Oldest removed when exceeded. Default: 1000

Memory Operations

store (topic: "store")
{
  "topic": "store",
  "payload": {
    "content": "Memory content",
    "type": "optional type",
    "metadata": { "any": "data" }
  }
}

Output: {stored: true, memory: {...}}

search (topic: "search")
{
  "topic": "search",
  "payload": "search term"
}

Output: {results: [...], count: n}

count (topic: "count")

Output: {count: n}

get all (no topic)

Output: {memories: [...]}

Details

⚠️ Volatile Storage: All memories are lost when Node-RED restarts. Use File Memory for persistence.

When to Use

  • Session-only conversation context
  • High-performance scenarios
  • Development/testing
  • Temporary caching

Best Practices

  • Set appropriate maxItems to prevent memory bloat
  • Use File Memory if persistence is needed
  • Clear memories periodically for long-running sessions