AI Assistant with RAG, MCP, and Search Engine Capabilities

This n8n workflow creates a powerful AI assistant that can answer questions using a combination of Retrieval-Augmented Generation (RAG) via the Model Context Protocol (MCP) and a search engine tool. It's designed to provide comprehensive and context-aware responses to user queries.

What it does

This workflow automates the process of answering user questions by:

1. Listening for chat messages: It triggers whenever a new chat message is received, acting as the entry point for user queries.
2. Maintaining conversational context: It uses a simple memory buffer to keep track of the conversation history, allowing the AI to understand the ongoing dialogue.
3. Leveraging an AI Agent: An AI Agent orchestrates the process, deciding whether to use a search engine or the Model Context Protocol (MCP) based on the user's query and the available tools.
4. Utilizing a Search Engine (via MCP Client Tool): When a search is needed, the AI Agent activates the MCP Client Tool, which can be configured to interact with a search engine to retrieve relevant information.
5. Utilizing Retrieval-Augmented Generation (RAG) (via MCP Client Tool): For questions requiring knowledge from a specific context, the AI Agent uses the MCP Client Tool to perform RAG, fetching information from a defined knowledge base.
6. Generating responses with an OpenAI Chat Model: The core of the AI's intelligence is powered by an OpenAI Chat Model, which processes the user's query, conversational context, and any retrieved information to formulate a coherent and helpful response.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance to host and execute the workflow.
• OpenAI API Key: For the "OpenAI Chat Model" node to generate AI responses.
• MCP Client Tool Configuration: The "MCP Client Tool" node requires configuration to connect to your Model Context Protocol (MCP) endpoint and potentially a search engine. This might involve setting up credentials or specific API endpoints within the node's settings.
• Chat Service Integration: The "When chat message received" trigger needs to be connected to a chat service (e.g., Slack, Telegram, Discord, custom chat UI) to receive user messages.

Setup/Usage

1. Import the workflow: Download the JSON provided and import it into your n8n instance.
2. Configure Credentials:
   • OpenAI Chat Model: Provide your OpenAI API Key in the credentials section for this node.
   • MCP Client Tool: Configure the MCP Client Tool with the necessary API keys, endpoints, or other details to connect to your MCP instance and any integrated search engine.
3. Configure Chat Trigger: Set up the "When chat message received" trigger to listen for messages from your desired chat platform.
4. Activate the workflow: Once configured, activate the workflow to start processing chat messages and answering questions.