Back to Templates

Build a Scalable AI Chatbot with GPT-4 and Pipedream: Calendly, Gmail Integration

Created by

Created by: Davide || n3witalia

Davide

Last update

Last update 9 hours ago

Share


This workflow contains community nodes that are only compatible with the self-hosted version of n8n.

This n8n workflow integrates the powerful Pipedream MCP server with AI capabilities to create a smart, extensible assistant that can interact with over 2,700 APIs and 10,000+ tools — all within a secure and modular structure.

image

This setup seamlessly integrates Pipedream's MCP server with n8n, enabling your AI assistant to leverage thousands of APIs and tools securely.


Benefits

  • Massive Tool Access: Instantly connect 2,700+ APIs using Pipedream MCP tools — from productivity apps to custom APIs — with zero-code integration.
  • Dynamic AI Agent: The use of a LangChain agent allows for flexible tool execution and contextual conversations, powered by GPT.
  • Easy Customization: Simply copy your MCP tool URL into the respective sseEndpoint field to extend the agent’s capabilities.
  • Scalable and Modular: Add or remove tools (like Slack, Notion, Stripe, etc.) without altering the core logic.
  • Secure and Revocable: Credentials and API access can be managed directly via Pipedream’s MCP dashboard.

How It Works

  1. Chat Trigger: The workflow begins when a chat message is received via the When chat message received node, which acts as the entry point.
  2. AI Agent Processing: The message is passed to the AI Agent node, which orchestrates the interaction using the connected tools and memory.
  3. Language Model: The OpenAI Chat Model (GPT-4.1-mini) processes the user's input and generates responses or actions.
  4. Memory: The Simple Memory node retains context from the conversation to enable coherent multi-turn interactions.
  5. Tool Integration:
    • The Calendly and Gmail nodes (connected via Pipedream's MCP server) allow the AI to perform actions like scheduling events or sending emails. These tools use SSE (Server-Sent Events) endpoints provided by Pipedream.
  6. Response: The AI Agent combines the model's output and tool responses to deliver a final reply to the user.

Set Up Steps

  1. Sign Up for Pipedream:
    • Create an account on and set up your MCP server.
  2. Configure MCP Tools:
    • Connect your accounts (e.g., Calendly, Gmail) in Pipedream and obtain the SSE endpoints for each tool (e.g., https://mcp.pipedream.net/xxx/calendly_v2).
  3. Update n8n Nodes:
    • Replace the placeholder SSE endpoints in the Calendly and Gmail nodes with your Pipedream MCP URLs.
  4. OpenAI Credentials:
    • Ensure the OpenAI Chat Model node has valid API credentials (configured under "OpenAi account").
  5. Activate Workflow:
    • Enable the When chat message received node (currently disabled) and deploy the workflow.

Need help customizing?

Contact me for consulting and support or add me on Linkedin.