Turn a chat or voice assistant into a smart support agent that handles bookings, order lookups, and product questions in one place. Teams can route customer requests to the right action without switching tools.

There are three webhook paths. One path checks an order by calling a tracking API and sends back a clear status message. Another path books meetings by taking a human date from the message, using an AI model to format start and end times, and creating a Google Calendar event. The third path answers questions using documents from Google Drive. Files are embedded, loaded into a Qdrant vector store, and used by an AI agent to return helpful answers. Each branch sends a fast response through a webhook reply.

You will need a Google account for Drive and Calendar, an OpenAI API key, a Qdrant endpoint, and a chatbot tool that can send webhooks. After setup, support teams can cut booking time, reduce manual tracking checks, and share accurate answers from company files. Good fits include retail stores, service desks, and product teams that want faster replies and fewer errors.

• Three webhook endpoints to handle orders, appointments, and knowledge questions
• Google Calendar event creation with AI formatted start and end times
• Order status retrieval via HTTP request to a tracking API and instant webhook reply
• Document ingestion from Google Drive and vector storage in Qdrant
• OpenAI embeddings and chat models for search and natural language responses
• Structured output parser to ensure valid date fields for calendar
• Conversational agent that uses the vector store as a tool for accurate answers
• Webhook response nodes that return clear messages to the user

• Reduce booking time from 10 minutes to under 1 minute with automatic calendar creation
• Eliminate manual order lookups by routing requests to the tracking API
• Improve answer accuracy by using company files instead of memory
• Handle more requests at once with three parallel webhook paths
• Connect chat, calendar, files, and AI in one flow without switching apps

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Google Drive, Google Calendar, OpenAI, Qdrant and Voiceflow. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, open the Google Drive node. In the Credential to connect with menu, click Create new credential, choose Google Drive OAuth2, and follow the on screen steps. Use a Google account that can access the source folders.
4. Open the Google Calendar node and create a new Google Calendar OAuth2 credential. Pick the correct calendar from the dropdown once connected.
5. Open any OpenAI node and create a new OpenAI API Key credential. Paste your API key from your OpenAI account API page and save.
6. Open the Qdrant nodes and create a Qdrant API credential. Enter your Qdrant URL and API key from your Qdrant dashboard. Make sure the collection name matches the one used in the nodes.
7. Configure the Google Drive folder path in the Get folder node. Confirm the Download Files node lists the right files. The Embeddings and Vector Store nodes will load and index these files.
8. Set the tracking API URL in the HTTP Request node. Map the Order number and Email fields to the incoming webhook data as shown.
9. Connect your chatbot tool. In your Voiceflow project, create three captures for order, appointment, and knowledge. Paste each n8n webhook URL into the matching action so messages reach the correct path.
10. Test the appointment path. Send a date phrase and check the Structured Output Parser and AI model return start and end fields. Confirm the Google Calendar event is created on the right calendar.
11. Test the order path by posting a sample Order number. Verify the tracking API returns status and that the webhook reply shows the status text.
12. Test the knowledge path by asking a product question. Confirm documents are embedded into Qdrant and that the agent replies with content drawn from your files.
13. If dates fail to create events, check the AI output in the parser node and ensure the time zone is set on your calendar. If Qdrant search fails, confirm the collection name and API key match your server settings.