Flowise Integration

CortexDB provides a custom Flowise node for adding persistent memory to your visual AI workflows.

Installation

1. Install the CortexDB Flowise node:

   cd ~/.flowise/nodes
   npm install cortexdbai
   

2. Restart Flowise
3. The CortexDB Memory node appears in the Memory category

Configuration

Drag the CortexDB Memory node onto your canvas and configure:

| Field | Description | |---|---| | Base URL | CortexDB server URL (e.g., https://api.cortexdb.ai) | | API Key | Your CortexDB API key | | Tenant ID | Tenant identifier |

Usage

Connect the CortexDB Memory node to any LLM chain or agent node. The node automatically:

1. Recalls relevant context before the LLM processes the input
2. Stores the conversation turn after the LLM responds
3. Injects memory context into the system prompt

Example Flow

[Chat Input] → [CortexDB Memory] → [ChatOpenAI] → [Chat Output]

The memory node enriches the conversation with relevant historical context from CortexDB.

Under the Hood

The Flowise node calls the CortexDB REST API:

# Recall (before LLM processing)
curl -X POST https://api.cortexdb.ai/v1/recall \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{"query": "user input text", "tenant_id": "my-app"}'

# Remember (after LLM response)
curl -X POST https://api.cortexdb.ai/v1/remember \
  -H "Authorization: Bearer your-api-key" \
  -H "Content-Type: application/json" \
  -d '{"content": "conversation turn text", "tenant_id": "my-app"}'