N8N Tools - GraphRAG Agent & Tool

Complete GraphRAG integration with both AI Agent Node and AI Tool Node for maximum flexibility in your AI workflows.

About

This package includes two nodes for maximum flexibility:

🤖 GraphRAG Agent Node (Regular Node)

Traditional N8N node with input/output for direct workflow integration. Perfect for processing documents and building knowledge graphs in your workflows.

Features:

• Traditional N8N workflow integration
• Direct input/output data flow
• Complete pipeline processing
• Batch document processing

🔧 GraphRAG Tool Node (AI Tool)

AI Agent Tool that can be connected to LangChain AI Agents. Enables AI agents to dynamically use GraphRAG capabilities as tools during conversations and reasoning.

Features:

• LangChain AI Agent integration
• Dynamic tool calling by AI agents
• Conversational GraphRAG analysis
• Context-aware responses
• Real-time document analysis during agent conversations

Both nodes provide identical GraphRAG functionality but serve different use cases - direct workflow integration vs AI agent tool usage.

Features

• 🧠 Two Node Types: Agent Node + Tool Node for maximum flexibility
• 🔗 AI Agent Integration: Tool Node works with LangChain AI Agents
• 📄 Document Processing: Intelligent document chunking and entity extraction
• 🕸️ Knowledge Graphs: Graph construction, querying, and management
• 🎯 Vector Search: Semantic embeddings using Sentence Transformers
• 🔍 Hybrid Search: Combined graph + vector search capabilities
• 📊 Analysis & Insights: Automated pattern detection and insight generation
• ⚡ Complete Pipeline: End-to-end processing capabilities
• 🛠️ Real ML Models: NLTK, spaCy, scikit-learn, NetworkX, FAISS, ChromaDB

Installation

npm install n8n-nodes-n8ntools-graphrag-agent

Credentials

Both nodes use the same N8N Tools API credentials:

1. Sign up at n8ntools.io
2. Get your API Key from the dashboard
3. Create N8N Tools API credentials in N8N:
   • API URL: https://n8ntools.io
   • API Key: Your N8N Tools API key

Operations

Both nodes support the same GraphRAG operations:

GraphRAG Agent Node Operations

• Document Processing: Intelligent document chunking and entity extraction
• Graph Construction: Create and query knowledge graphs from document entities
• Vector Embedding: Generate semantic vector embeddings and similarity search
• Hybrid Search: Combined graph and vector search capabilities
• Analysis & Insights: Generate insights and detect patterns
• Complete Pipeline: End-to-end GraphRAG processing

GraphRAG Tool Node Operations

The Tool Node provides the same operations but optimized for AI Agent usage:

• Document Analysis: Analyze documents and extract structured information
• Knowledge Graph Query: Query knowledge graphs for entities and relationships
• Hybrid Search: Perform vector + graph hybrid search across documents
• Entity Extraction: Extract named entities and relationships from text
• Summary Generation: Generate intelligent summaries using graph context

Operation Modes

• Standard: Basic GraphRAG processing
• Advanced: Enhanced processing with multiple ML models
• Custom: Configurable processing parameters

Example Configuration

GraphRAG Agent Node (Regular Node)

{
  "resource": "completePipeline",
  "pipelineConfig": "standard",
  "inputData": "Your document content here",
  "query": "What insights can you provide?",
  "resultLimit": 5
}

GraphRAG Tool Node (AI Tool)

The Tool Node is configured differently and is designed for AI Agent integration:

{
  "name": "graphrag_analyzer",
  "description": "Analyze documents using GraphRAG - extract entities, build knowledge graphs, perform hybrid search, and provide intelligent insights from complex document collections.",
  "operationMode": "document_analysis",
  "inputSchema": {
    "type": "object",
    "properties": {
      "query": {
        "type": "string",
        "description": "The query or document text to analyze"
      },
      "operation": {
        "type": "string",
        "enum": ["analyze", "search", "extract", "summarize"],
        "description": "Specific operation to perform"
      }
    },
    "required": ["query"]
  }
}

AI Agent Usage Example

When connected to a LangChain AI Agent, the agent can call the GraphRAG Tool like this:

{
  "query": "Analyze this research paper about climate change",
  "operation": "analyze",
  "options": {
    "language": "en",
    "max_results": 10,
    "include_graph": true,
    "confidence_threshold": 0.7
  }
}

Architecture

The GraphRAG Agent uses a specialized architecture with real ML models:

• Text Processing: NLTK + spaCy for natural language processing
• Embeddings: Sentence Transformers for semantic understanding
• Graph Analysis: NetworkX for knowledge graph operations
• Vector Search: FAISS for fast similarity search
• Vector Storage: ChromaDB for optimized storage

Benefits

GraphRAG Agent Node

• Direct Integration: Traditional N8N workflow integration
• Batch Processing: Process multiple documents in workflows
• Data Pipeline: Part of larger data processing workflows
• Scheduled Processing: Automated document analysis on schedule

GraphRAG Tool Node

• AI Agent Integration: Works seamlessly with LangChain AI Agents
• Conversational AI: Enables AI agents to analyze documents during conversations
• Dynamic Processing: Real-time analysis based on user queries
• Context-Aware: Maintains conversation context for better responses

Both Nodes

• Privacy: 100% secure processing through N8N Tools API
• Performance: Real ML models optimized for speed
• Cost-Effective: Transparent pricing with N8N Tools credits
• Integration: Built specifically for N8N ecosystem

Support

• Documentation: N8N Tools Docs
• Issues: GitHub Issues

License

MIT License - see LICENSE file for details.