Automate Movie Recommendations with n8n, OpenAI, MongoDB

What This Automation Does

This workflow listens for chat messages that ask for movie suggestions.
It turns those messages into special database queries to find movies with certain ratings or genres.
Users can also save favorite movies to the database from the chat.
This saves time and errors from searching movies by hand.
The result is quick, tailored movie recommendations with easy favorite saves.

Who Should Use This Workflow

This workflow is good for cinema club coordinators and movie fans who want to cut down on movie search time.
It fits anyone wanting fast, AI-powered movie suggestions from their own database.
You don’t need to be a programmer but must have basic n8n skills.

Tools and Services Used

• n8n Chat Trigger node: Catches live chat messages from users.
• OpenAI Chat Model: Understands chat input and creates proper database queries.
• MongoDBAggregate node: Runs queries on the movies collection in MongoDB.
• Window Buffer Memory: Keeps last 10 chats to help AI remember context.
• AI Agent – Movie Recommendation: Controls chat response flow and talking to MongoDB.
• Insert Favorite Workflow node: Adds saved favorite movies back into MongoDB.

Workflow Inputs, Process, and Output

Inputs

• User sends a chat message with a movie request to the webhook URL.

Processing Steps

• Webhook node receives chat message.
• OpenAI Chat Model reads message and makes a MongoDB aggregation query.
• Window Buffer Memory keeps recent chats so AI keeps conversation flow.
• MongoDBAggregate node runs the AI-created query against movies collection.
• AI Agent – Movie Recommendation guides workflow, decides when to add favorites or respond.
• Insert Favorite Workflow node adds movies into user favorites if requested.

Outputs

• Returns a list of recommended movies showing data like title and rating.
• Favorite movies get saved to the database automatically when chosen.
• Maintains chat history for smooth, natural conversation.

Beginner Step-by-Step: How to Use This Workflow in Production

Step 1: Download and Import the Workflow

1. Find the Download button on this page and save the workflow file to your computer.
2. Open your n8n editor where you want to run the workflow.
3. Click the menu and select Import from File.
4. Choose the saved workflow file and load it into your n8n instance.

Step 2: Configure Credentials and IDs

1. Go to each node that needs credentials, like OpenAI Chat Model and MongoDBAggregate.
2. Add your OpenAI API Key in the credential settings.
3. Enter your MongoDB credentials securely in the credential area.
4. If you use specific IDs, emails, or collection names, update those nodes accordingly.

Step 3: Review and Adjust Prompts or Queries if Needed

1. Check if the MongoDB pipeline expression and AI prompts fit your movie database fields.
2. Adjust the prompt text or sample pipeline if your data has different structure.

{{ $fromAI("pipeline", "The MongoDB pipeline to execute" , "string" , [{"$match" : { "rating" : 5 } }])}}

Step 4: Test the Workflow

1. Send a chat message to the webhook URL. Example: “Recommend top-rated Western movies.”
2. Check that the response gives movie suggestions as expected.
3. Try saving a favorite to confirm inserts work as planned.

Step 5: Activate the Workflow for Production Use

1. Once tests pass, toggle the workflow to active in n8n.
2. Make sure the webhook URL can be reached by your users or systems sending chats.
3. Monitor workflow executions and logs for any issues.
4. Consider deploying on more secure infrastructure or self-host n8n for privacy and control.

Customizations

• Change AI prompt in AI Agent – Movie Recommendation node to add filters by year, director, or actor.
• Adjust Window Buffer Memory context length to remember fewer or more chat messages.
• Secure your webhook by restricting allowed origins or adding tokens for access control.
• Expand favorite insert data to include ratings or user comments in the Insert Favorite Workflow.
• Add new nodes or tools to call other APIs like TMDb for live movie data.

Troubleshooting

• MongoDBAggregate node fails due to pipeline syntax error.
  The AI-generated pipeline may not fit MongoDB syntax or version.
  Test the pipeline directly with MongoDB tools or simplify the AI prompt.
• OpenAI Chat Model node shows authentication errors.
  Verify the OpenAI API Key in n8n credentials is correct and active.
• Chat webhook does not trigger workflow.
  Check the webhook URL, allowed origins, and public access settings on the Webhook node.

Pre-Production Checklist

• Run test queries on MongoDBAggregate node to make sure it connects and fetches data.
• Check OpenAI Chat Model calls with sample messages to confirm no API errors.
• Verify Window Buffer Memory holds enough message context for smooth answers.
• Send test chats to webhook and watch for expected replies.
• Check that the favorite insert has correct data format before going live.
• Backup MongoDB movies collection before enabling user inserts.

Deployment Guide

Turn the workflow active in n8n to start receiving chat messages.
Make sure the webhook is reachable depending on user location or network.
Watch workflow executions under the n8n interface to see errors or results.
Set up alerts or logs if necessary to quickly spot problems.
Consider running n8n on your own server for control over data and usage.
Self-host n8n options help with privacy and cost control.

Summary

✓ Saves time by automating movie searches from chat.
✓ Uses AI to create custom database queries based on user requests.
✓ Provides fast, personalized movie recommendations.
✓ Lets users save favorites directly via chat.
✓ Maintains conversation flow with chat memory.
✓ Easily set up inside n8n with minimal coding.
→ Good for any movie fan or club wanting quick movie picks and easy favorite management.