n8n-nodes-cloudflare-vectorize

A comprehensive n8n community node for Cloudflare Vectorize, enabling vector database operations for semantic search, embeddings management, and RAG (Retrieval-Augmented Generation) applications.

Features

• 🚀 Complete Vector Operations: Insert, upsert, query, delete vectors
• 🔍 Advanced Search: Semantic similarity search with configurable parameters
• 📊 Index Management: Create, configure, and manage vector indexes
• 🏷️ Metadata Support: Rich metadata filtering and indexing
• 🎯 AI Agent Ready: Optimized for use with n8n AI Agent and MCP Trigger
• 📝 Namespace Support: Organize vectors with namespace segmentation
• ⚡ Batch Operations: Efficient bulk vector processing
• 🛠️ Developer Friendly: Full TypeScript support with comprehensive types

Installation

From n8n Community Nodes (Recommended)

1. Go to Settings > Community Nodes in your n8n instance
2. Click Install a community node
3. Enter: n8n-nodes-cloudflare-vectorize
4. Click Install

Manual Installation

# In your n8n installation directory
npm install n8n-nodes-cloudflare-vectorize

# Restart n8n

Quick Start

1. Set up Cloudflare Vectorize

1. Create a Cloudflare account
2. Create a vector index using Wrangler CLI:

   npx wrangler vectorize create my-index --dimensions=1536 --metric=cosine
   

3. Get your Account ID from the Cloudflare dashboard
4. Create an API token with Vectorize permissions

2. Configure Credentials in n8n

1. Add new credentials: Cloudflare Vectorize API
2. Enter your:
   • Account ID: Found in your Cloudflare dashboard
   • API Token: Created with Vectorize read/write permissions
   • API Endpoint: Leave default unless using custom endpoint

3. Use in Workflows

Add the Cloudflare Vectorize node to your workflow and select from these resources:

• Index: Manage vector indexes
• Vector: Insert, query, and manage vectors
• Metadata: Configure metadata indexes for filtering
• Utility: List vectors and get index information

Usage Examples

Insert Vectors

{
  "resource": "vector",
  "operation": "insert",
  "indexName": "my-embeddings",
  "vectors": [
    {
      "id": "doc1",
      "values": [0.1, 0.2, 0.3, ...],
      "metadata": {
        "title": "Document 1",
        "category": "tech"
      }
    }
  ]
}

Query Similar Vectors

{
  "resource": "vector",
  "operation": "query",
  "indexName": "my-embeddings",
  "queryVector": [0.1, 0.2, 0.3, ...],
  "topK": 5,
  "returnValues": true,
  "returnMetadata": "all"
}

Create Index

{
  "resource": "index",
  "operation": "create",
  "indexName": "new-index",
  "dimensions": 1536,
  "metric": "cosine",
  "description": "Embeddings for documents"
}

AI Agent Integration

This node is optimized for use with n8n's AI Agent and MCP Trigger:

{
  "tool": "cloudflare-vectorize",
  "operation": "query-similar-vectors",
  "parameters": {
    "query": "Find documents about machine learning",
    "index": "knowledge-base",
    "limit": 3
  }
}

Operations Reference

Index Resource

Vector Resource

Metadata Resource

Utility Resource

Configuration

Vector Formats

Vectors can be provided as:

• Array of numbers: [1.0, 2.0, 3.0]
• Float32Array or Float64Array
• n8n expressions: {{ $json.embedding }}

Distance Metrics

• cosine: Cosine similarity (recommended for most use cases)
• euclidean: Euclidean distance
• dot-product: Dot product similarity

Metadata Filtering

Create metadata indexes to enable filtering:

{
  "resource": "metadata",
  "operation": "createIndex",
  "indexName": "my-index",
  "propertyName": "category",
  "type": "string"
}

Then use in queries:

{
  "resource": "vector",
  "operation": "query",
  "filter": {
    "category": "technology"
  }
}

Error Handling

The node provides detailed error messages for common issues:

• Authentication errors: Check API token permissions
• Index not found: Verify index name and existence
• Dimension mismatches: Ensure vector dimensions match index
• Rate limiting: Automatic retry with exponential backoff
• Network errors: Configurable retry logic

Limits

Cloudflare Vectorize has the following limits:

• Vectors per request: Up to 1,000 for inserts/upserts
• Query topK: Up to 100 (20 with returnValues/returnMetadata)
• Metadata indexes: Up to 10 per index
• Vector dimensions: Up to 1536
• Namespace length: Up to 64 characters
• Metadata size: Up to 10KB per vector

Development

Building from Source

git clone https://github.com/jezweb/n8n-nodes-cloudflare-vectorize.git
cd n8n-nodes-cloudflare-vectorize
npm install
npm run build

Testing Locally

# Link to local n8n
npm link
cd ~/.n8n/custom
npm link n8n-nodes-cloudflare-vectorize

# Restart n8n

Contributing

Contributions are welcome! Please read our contributing guidelines and submit issues and pull requests on GitHub.

License

MIT License - see LICENSE file for details.

Support

• 📖 Cloudflare Vectorize Documentation
• 🐛 Report Issues
• 💬 n8n Community
• 📧 Email Support

Related

• n8n-nodes-cloudflare-d1 - Cloudflare D1 database integration
• Cloudflare Vectorize - Official documentation
• n8n - Workflow automation platform