AI & researchAgentChat TriggerDocument Default Data LoaderEmbeddings Open AiLM Chat Google Gemini

Travel AssistantAgent

Configured with Agent, Chat Trigger, Document Default Data Loader, Embeddings Open Ai, and additional nodes. Review the connections and setup before adapting this template.

Download workflow JSON View files on GitHub ↗11 nodes · MIT

Workflow preview

Node categories in this workflow

Trigger
Data store
AI & models
Files / media
Transform

Icons describe general node roles, inferred from node types. Inspect the parameters for the configured action.

Drag to pan; use + and − to zoom. Reconstructed from the exported node positions; nothing runs here. Dashed lines represent non-main connections. Full node names and notes appear below.

How it works

The exported workflow contains the following configured nodes. Their connections are shown above; inspect the JSON for the complete logic, expressions, and external actions.

Workflow node names, types, and operations
Node	Category	Type	Operation
When chat message received	Trigger	Chat Trigger	See node parameters
MongoDB Chat Memory	Data store	Memory Mongo Db Chat	See node parameters
Google Gemini Chat Model	AI & models	LM Chat Google Gemini	See node parameters
MongoDB Atlas Vector Store	Data store	Vector Store Mongo Dbatlas	See node parameters
Embeddings OpenAI	AI & models	Embeddings Open Ai	See node parameters
AI Traveling Planner Agent	AI & models	Agent	See node parameters
Webhook	Trigger	Webhook	See node parameters
Default Data Loader	Files / media	Document Default Data Loader	See node parameters
Recursive Character Text Splitter	Transform	Text Splitter Recursive Character Text Splitter	See node parameters
MongoDB Atlas Vector Store1	Data store	Vector Store Mongo Dbatlas	See node parameters
Embeddings OpenAI1	AI & models	Embeddings Open Ai	See node parameters

Setup

Download the JSON and import it into an isolated n8n instance. Leave the workflow inactive.
Connect your own credentials through n8n’s credential store. Replace REPLACE_WITH_* values with your resource identifiers.
Inspect API destinations, Code nodes, database queries, file paths, and outbound actions.
Run with synthetic data and harmless destinations. Enable triggers only after checking the actual outputs.

Requirements

An n8n instance with the listed node types and compatible versions. Service-connected workflows may require accounts, credentials, or paid APIs.

Credential types declared in the source

googlePalmApi
mongoDb
openAiApi

Workflow notes

AI Traveling Agent Powered by MongoDB Atlas for Memory and vector search.

Atlas MongoDB Memory Node

The memory node allows the agent to persist and retrieve conversation based on threads in the database. It uses MongoDB felxible store capabilities to allow different type of threads and messages (Image, audio, video etc.) to be stored easily and effectivley

Atlas MongoDB Vector Store Node

Atlas Vector Store tool allows the agent to get up to date points of interest from our vector store database populated and embedded with OpenAI Embeddings.

You will need to:

Setup your Google API Credentials for the Gemini LLM
Setup your OpenAI Credentials for the OpenAI embedding nodes.
MongoDB Atlas project and Cluster. Get a hold of the connection string and make sure to have your IP Access list enabled (for ease of testing try 0.0.0.0/0 access.
Setup you MongoDB Credentials account with the correct connection string and database name.
Vector Search Tool - uses Atlas Vector Search index you will create on your database for the points_of_interest collection:

// index name : "vector_index"
// If you change an embedding provider make sure the numDimensions correspond to the model.
{
  "fields": [
    {
      "type": "vector",
      "path": "embedding",
      "numDimensions": 1536,
      "similarity": "cosine"
    }
  ]
}

Once all of that is configured you will need to send the loading webhook with some data points (see example).

This should create vectorised data in points_of_interest collection.

Once you have data points there try to ask the Agent questions about the data points and test the response. Eg. "Where should I go for a romantic getaway?"

Additional Resources

CURL Command to Ingest Data.

Here is an example of how you can load data into your webhook once its active and ready to get requests.

curl -X POST "https://<account>.app.n8n.cloud/webhook-test/ingestData" \
  -H "Content-Type: application/json" \
  -d '{
    "raw_body": {
      "point_of_interest": {
        "title": "Eiffel Tower",
        "description": "Iconic iron lattice tower located on the Champ de Mars in Paris, France."
      }
    }
  }'

Vector Search data ingestion

Using webhook to ingest data to the MongoDB points_of_interest collection.

This can be done in other ways like loading from wbesites/git/files or other supported data sources.

Travel AssistantAgent

Purpose and structure

This folder contains the Travel AssistantAgent workflow. Its executable nodes use agent, chatTrigger, documentDefaultDataLoader, embeddingsOpenAi, lmChatGoogleGemini, memoryMongoDbChat, textSplitterRecursiveCharacterTextSplitter, vectorStoreMongoDBAtlas, webhook. The table and connection map below are derived from the JSON, not an inferred guarantee of behavior.

Setup

Import inactive. Configure your own credentials and replace REPLACE_WITH_* values. Review external destinations and Code nodes before running.

Entry points

When chat message received, Webhook

Verification

Sanitized source candidate; not runtime-tested. This overview is generated from node metadata. Read the folder README for the connection map, credential types, and license.

Attribution

https://github.com/Zie619/n8n-workflows/blob/94007c1445d9258a7da116646b79473e7c7c3282/workflows/Stickynote/2039_Stickynote_Webhook_Automation_Webhook.json MIT; preserve source notices.

More in AI & research

"Hey Siri, Ask Agent" workflow #️⃣Nostr #damus AI Powered Reporting + Gmail + Telegram (G) - Email Classification