Modular AI Command Center for Google Workspace and RAG - n8n Workflow

AI Developers & Integrators: Those looking to build complex, tool-using AI systems using n8n.
Business Operations Managers: Users needing automated, natural language control over spreadsheets (Google Sheets).
RAG System Architects: Professionals requiring a complete data ingestion pipeline (Google Drive to Supabase Vector DB) within an n8n workflow.
Multi-Channel Support Teams: Teams who need a unified AI interface for Slack, Telegram, Gmail, and WhatsApp.

This comprehensive n8n workflow establishes a sophisticated 'AI Command Center' designed for enterprise operations. It solves the challenge of siloed data and limited AI capabilities by providing specialized, tool-equipped AI agents capable of performing complex tasks across multiple systems.

The core value lies in its modularity: it uses the Model Context Protocol (MCP) to manage Google Sheets and Google Drive operations via natural language. Users can initiate requests from channels like Slack or Gmail, and the central n8n agent determines the correct sequence of actions, which might include reading data from a vector database (RAG), updating a Google Sheet, or generating a structured report. This complex n8n template provides a blueprint for scalable, intelligent business automation.

The entire system is orchestrated by the MAIN AGENT, which uses a Reasoning Model (OpenRouter) and Postgres Chat Memory to maintain context across interactions.

1. Incoming Request: The n8n trigger initiates from a communication channel (Gmail Trigger, Slack Trigger, Telegram Trigger, or WhatsApp Trigger). The message content is passed to the MAIN AGENT.


2. Agent Reasoning: The MAIN AGENT analyzes the request and decides which specialized tool (or sub-agent) is required. Available tools include MCP Sheets (for Google Sheets), Knowledge Agent1 (for external data/RAG), Linkedin Scraper, and Create reports.


3. Data Interaction (MCP): If the request involves Google Sheets (e.g., "read Q3 sales data"), the agent calls the MCP Sheets client. This client triggers the dedicated Google Sheets MCP server n8n workflow, which executes the specific spreadsheet n8n node (e.g., Google Sheets - Read Data). The same modular approach is used for Google Drive file search and reading.


4. Knowledge Retrieval (RAG): The Knowledge Agent1 handles requests requiring general or document-specific information. It leverages the General knowledge1 Supabase Vector Store, using OpenAI embeddings and Cohere Reranking for high-quality retrieval. It can also access external search via Message a model in Perplexity1.


5. Data Ingestion Pipeline: A separate n8n workflow path (When clicking ‘Execute workflow’) handles RAG population. It downloads files from Google Drive, processes them using a Default Data Loader and Recursive Character Text Splitter, creates embeddings via the Embeddings OpenAI1 n8n node, and uploads the chunks to the Add to Supabase Vector DB.


6. Output: Once the MAIN AGENT has the answer, it uses the appropriate output n8n node (Gmail, Telegram, Slack) to send the structured response back to the user.

To deploy this comprehensive n8n workflow, follow these steps:

1. Import the n8n Workflow: Copy the provided JSON data and paste it directly into your n8n instance using the 'New' -> 'Import from JSON' option.


2. Configure Credentials: This n8n template requires numerous credentials. You must set up connections for:

1. Establish Inter-Workflow Links (MCP): Since this n8n workflow uses the Model Context Protocol (MCP), you must ensure that the client nodes (MCP Sheets, search about any doc in google drive1) are correctly configured to point to the server webhook IDs generated by the corresponding MCP Server nodes in this same n8n workflow structure (MCP Server Sheets, Google Drive MCP Server).


2. Activate Triggers: Ensure the multi-channel n8n trigger nodes (Gmail, Slack, Telegram, WhatsApp) are set up with correct webhooks or polling configurations and then activate the entire n8n workflow.

This advanced n8n workflow utilizes dozens of n8n nodes for complex orchestration:

MAIN AGENT (n8n node): The core intelligence, using the 'Reasoning model (recommended)' (OpenRouter) as its brain. It handles tool selection and conversational flow, leveraging Postgres Chat Memory.
MCP Server Sheets (n8n trigger): A dedicated n8n trigger for Google Sheets operations, allowing the AI to call six distinct Sheets operations (Read Data, Add Data, Update Data, etc.) via natural language.
Google Sheets - Read Data (n8n node): Used by the MCP server to retrieve spreadsheet data. Key Configuration: AI parameters like DocumentID, SheetName, and Range are exposed to the AI agent.
Google Drive MCP Server (n8n trigger): Routes requests for file search and reading to the file management sub-workflows. The Read File From GDrive sub-workflow handles complex file type routing.
Embeddings OpenAI1 (n8n node): Critical for the RAG pipeline. It converts processed documents into vector embeddings for storage in the Add to Supabase Vector DB.
Recursive Character Text Splitter1 (n8n node): Prepares large documents for embedding by breaking them into manageable chunks, ensuring optimal knowledge retrieval.
Create structured Reports (n8n agent): A specialized sub-agent that uses the OpenRouter Chat Model to format final answers into polished reports, which are then converted using the Markdown to HTML n8n node and sent via Gmail.
Gmail Trigger / Slack Trigger / Telegram Trigger (n8n trigger): The entry points for user interaction across different communication platforms. They pass incoming messages to the Set message variable n8n node before reaching the main agent.