Smart Gmail Cleaner with AI Validator & Telegram Alerts

This n8n workflow automates the process of cleaning up your Gmail inbox by identifying and processing specific emails, validating their content using AI, and sending alerts via Telegram. It's designed to help you manage your inbox more efficiently by automatically handling emails that meet certain criteria.

What it does

This workflow performs the following key steps:

1. Manual Trigger: The workflow is initiated manually, allowing you to run it on demand.
2. Gmail Email Retrieval: It connects to your Gmail account to retrieve emails. (The specific criteria for email retrieval are not defined in the provided JSON, but would typically be configured in the Gmail node).
3. AI Content Validation: The retrieved email content is passed to an AI Agent (likely a Large Language Model like Google Gemini) for analysis and validation. This step likely checks if the email content meets certain criteria for deletion or further action.
4. Structured Output Parsing: The AI Agent's response is then parsed to extract structured information, ensuring that the output is consistent and usable for subsequent steps.
5. Conditional Logic (If Statement): Based on the AI validation, the workflow uses an "If" node to route the email.
   • TRUE Branch: If the AI validation returns true (indicating the email should be processed), the workflow proceeds to edit fields and send a Telegram alert.
   • FALSE Branch: If the AI validation returns false (indicating the email should not be processed), the workflow does nothing further for that email.
6. Edit Fields (Set): For emails that pass the AI validation, this node likely modifies or adds specific data points related to the email, preparing it for the next action.
7. Telegram Alert: A message is sent to a specified Telegram chat, alerting you about the processed email. This could include details about the email and the action taken.
8. Aggregate (Optional): The workflow includes an "Aggregate" node, which could be used to combine or summarize data from multiple processed emails before a final action, though its specific use case here is not fully detailed by the JSON.
9. No Operation: The "No Operation" node on the false branch simply acts as a placeholder, indicating that no action is taken for emails that do not meet the AI validation criteria.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running instance of n8n.
• Gmail Account: Credentials for a Gmail account to access your emails.
• Google Gemini (or similar LLM) API Key/Access: Credentials for an AI service (like Google Gemini) to power the AI Agent and Chat Model nodes.
• Telegram Account & Bot Token: A Telegram bot and its token, along with the chat ID where you want to receive alerts.

Setup/Usage

1. Import the Workflow: Import the provided JSON into your n8n instance.
2. Configure Credentials:
   • Set up your Gmail credentials (OAuth 2.0 is recommended).
   • Configure your Google Gemini Chat Model credentials (API key or service account).
   • Set up your Telegram credentials (Bot Token).
3. Customize Nodes:
   • Gmail Node: Adjust the "Gmail" node to specify which emails to retrieve (e.g., search queries, labels, unread status).
   • AI Agent Node: Configure the "AI Agent" node with the specific instructions or prompts for validating email content. This is crucial for defining what constitutes a "cleanable" email.
   • Structured Output Parser: Ensure the parser is configured to correctly extract the validation result from the AI Agent's output.
   • If Node: Verify the condition in the "If" node to correctly evaluate the AI validation result.
   • Edit Fields (Set) Node: Customize the fields you want to set or modify based on the email content.
   • Telegram Node: Specify the chat ID and customize the message content for the alerts.
4. Execute Workflow: Run the workflow manually using the "When clicking ‘Execute workflow’" trigger to test its functionality. Once satisfied, you can set it up to run on a schedule or via a webhook for continuous automation.