Multi-Modal Personal Assistant Bot for Productivity - n8n Workflow

• Executives, managers, and entrepreneurs who need a proactive, witty, and highly capable personal AI assistant.


• Users looking for a complex n8n template demonstrating the full power of AI agents and external tool integration.


• Developers or automators seeking an advanced n8n node structure for conversational AI using Google Gemini.


• Anyone who wants to consolidate task management, email, calendar, and research into a single Telegram bot interface.

This sophisticated n8n workflow transforms a basic Telegram bot into a hyper-efficient personal AI assistant named 'Main,' characterized by a sharp, witty persona (like Donna Paulsen from Suits). The automation is designed to handle multi-modal inputs (text, voice, images) and orchestrate actions across several productivity tools—Gmail, Google Calendar, Todoist, Google Sheets, and Airtable—all driven by Google Gemini models.

The core value of this n8n workflow is its ability to contextually route complex requests, manage conversations via memory buffers, and perform scheduled proactive actions (like daily briefings and weekly team analysis). This n8n solution is far more than a simple chatbot; it is a complete digital twin managing your professional life, proving the versatility and power of the n8n automation platform.

The process begins with one of three primary n8n trigger types: a user input via the Telegram Trigger, a daily Schedule Trigger, or a weekly Schedule Trigger.

1. Input Handling & Pre-processing: When a message arrives via the Telegram Trigger, an n8n node called Switch determines the type of input: plain text, voice recording, or image/photo. Voice recordings are sent to Gemini for transcription; images are converted to Base64 via a Code n8n node and then analyzed by the Analyze image Gemini n8n node. Text is passed through directly.


2. Input Consolidation: All inputs are merged into a single stream using a Merge n8n node.


3. The Manager Agent: The consolidated input feeds into the central Manager Agent (an n8n LangChain agent node). This node, powered by a specific system prompt emphasizing a witty persona, decides which specialized sub-agent (tool) is required to fulfill the request. Conversation history is maintained using a Window Buffer Memory n8n node.


4. Tool Execution: The Manager Agent routes tasks to six underlying specialized AI Agent tools:

1. Output: Once the tool returns the necessary information or confirmation, the Manager Agent formulates the final, personalized, and character-driven response, which is then sent back to the user using the final Telegram n8n node.

To deploy this comprehensive n8n workflow template, follow these steps:

1. Import the n8n Workflow: Copy the provided JSON data and import it into your n8n instance via the 'Workflows' tab using the 'New' button -> 'Import from JSON'.


2. Credentials Setup: This n8n workflow requires credentials for multiple services. You must establish connections for:

1. Configuration: Update the hardcoded chatId variables in the Telegram nodes (e.g., =1235543129) to match your personal Telegram chat ID for two-way communication.


2. Activate: Ensure all required credentials are set up. Activate the n8n workflow using the toggle switch. The scheduled n8n trigger nodes will begin operating automatically.

Telegram Trigger (n8n trigger): Starts the n8n workflow upon receiving a message (text, photo, voice) from the connected chat.
Switch (n8n node): Essential flow control, routing the incoming Telegram message based on content type (voice, photo, text, or document) to the appropriate pre-processing path.
Analyze image & Transcribe a recording (Gemini n8n nodes): These nodes preprocess multi-modal data, turning images into text descriptions and voice files into transcribed text, making them legible for the main AI agent.
Manager Agent (n8n Agent Node): The core intelligence. This n8n node utilizes Google Gemini to process complex requests, maintain the witty persona, and dispatch tasks to specialized tools based on the user's intent. Utilizes Window Buffer Memory for context.
memorybase (Airtable n8n Tool): This agent tool handles all persistent data, enabling the AI to retrieve and store personal details, contacts, and past conversations using Airtable operations (Create, Search, Update, Delete).
emailagent (Gmail & Google Sheets n8n Tools): Allows the AI to perform email actions (Send, Draft, Reply, Label, Mark Unread) and look up contact details via a Google Sheets n8n node.
todoandtask_manager (Todoist n8n Tools): Manages personal tasks via Todoist, allowing the agent to create, update, move, delete, and close tasks.
Schedule Trigger (n8n trigger): Executes specific prompts (Edit Fields3 and Edit Fields4) at scheduled times (daily and weekly) to prompt the Manager Agent for proactive routines like daily briefings or team follow-ups.

• Telegram (n8n node): The final output node, sending the Manager Agent's generated response back to the user via Telegram.