n8n-nodes-elastic-advance-retrieve

This is a community node for n8n. It allows you to use Elasticsearch directly in your n8n workflows, executing advanced queries to retrieve documents.

Elasticsearch is a distributed search and analytics engine designed to execute complex queries on large volumes of data quickly and scalably.

n8n is a workflow automation platform with a fair-code license.

Installation
Operations
Credentials
Compatibility
Usage
Semantic Search
Resources

Installation

Follow the installation guide in the n8n community nodes documentation.

Operations

This node is implemented as a subnode and supports the following operations:

• Document Search: Allows you to perform advanced queries in Elasticsearch using the Elasticsearch DSL (Domain Specific Language) query format in JSON.

This node is specifically designed to be used as a tool by the AI Agent in n8n, allowing it to retrieve contextual information from Elasticsearch to enrich the generated responses.

Credentials

To authenticate with Elasticsearch, this node offers three authentication methods:

Basic Authentication (username/password)

• Host: URL of your Elasticsearch instance (e.g., http://localhost:9200)
• Username: Your Elasticsearch username
• Password: Your Elasticsearch password

API Key Authentication

• Host: URL of your Elasticsearch instance
• API Key: The API key generated from Elasticsearch

No Authentication

• Host: URL of your Elasticsearch instance (for instances that don't require authentication)

Note: You may need to configure your Elasticsearch instance to allow access via the credentials you choose.

Compatibility

• Requires n8n version 1.0.0 or higher
• Compatible with Elasticsearch 7.x and 8.x

Usage

This node allows you to perform advanced queries in Elasticsearch. Here are examples of how to use it:

Basic query to find all documents

{
  "query": {
    "match_all": {}
  }
}

Full-text query

{
  "query": {
    "match": {
      "field": "value to search"
    }
  }
}

More complex query with filters

{
  "query": {
    "bool": {
      "must": [
        { "match": { "title": "search" } }
      ],
      "filter": [
        { "term": { "status": "active" } },
        { "range": { "date": { "gte": "2023-01-01" } } }
      ]
    }
  },
  "sort": [
    { "date": { "order": "desc" } }
  ],
  "size": 20
}

The node will return the documents found as a result of the query, each with its metadata (_id, _score) and fields.

Semantic Search

This node is designed to make use of the semantic_text functionality, which allows Elasticsearch to retrieve information both semantically and lexically. In this version, the node only retrieves information but implements hybrid search capabilities.

Hybrid Search (semantic and lexical)

Hybrid search combines:

• Semantic search: Finds documents based on meaning and context, not just on exact word matches.
• Lexical search: Finds documents based on exact or partial text matches.

Example of a hybrid query in Elasticsearch:

{
  "size": 3,
  "_source": ["answer","ask"],
  "retriever": {
    "rrf": {
      "retrievers": [
        {
          "standard": {
            "query": {
              "semantic": {
                "field": "answer_semantic",
                "query": "How to obtain and activate the electronic signature?"
              }
            }
          }
        },
        {
          "standard": {
            "query": {
              "match": {
                "ask": {
                  "query": "eleectronic signature"
                }
              }
            }
          }
        }
      ]
    }
  }
}

This approach allows the AI Agent to obtain more relevant and contextual results based on the user's intent, rather than being limited to exact keyword matches.

Resources

• n8n Community Nodes Documentation
• Elasticsearch Documentation
• Elasticsearch DSL Query Guide
• Semantic Search with Elasticsearch

This node was made possible thanks to the incredible work of the Elasticsearch team and their continued efforts to make powerful search and analytics accessible to everyone.

Developed with care and shared with the n8n community by Alejandro Sánchez, in the spirit of open collaboration and curiosity-driven automation.