WhatsApp expense tracker with PostgreSQL database & AI-powered reports
Track Personal Finances with WhatsApp and AI Assistant Transform your WhatsApp into a powerful personal finance command center. This AI-powered workflow converts natural language messages into structured financial data, automates record-keeping, and delivers instant insights—all within your favorite messaging app. Who is this for? This template is perfect for: Personal finance enthusiasts who want effortless expense tracking Small business owners managing personal and business expenses Freelancers tracking income and expenses across projects Anyone who prefers messaging over complex finance apps Users seeking privacy with self-hosted financial data What problem is this workflow solving? Traditional expense tracking requires switching between apps, manual data entry, and complex spreadsheets. Most people abandon these systems within weeks. This workflow solves the friction by: Eliminating app-switching—everything happens in WhatsApp Converting natural language to structured data automatically Providing instant confirmations and reports Requiring zero learning curve or behavior change What this workflow does Smart Transaction Processing Send natural messages like Spent 300 on groceries at Walmart and the AI automatically extracts: Date: Today's date (or specified date) Category: Groceries Type: Expense/Income/Debt Amount: 300 Person/Company: Walmart Intelligent Message Classification The workflow automatically routes messages to three processing branches: Branch 1: Reports and analytics (show March expenses) Branch 2: Transaction logging (spent 50 on coffee) Branch 3: General financial chat (how can I save money?) Advanced Reporting Generate instant reports by messaging: today's report → Daily income/expense summary March vs April report → Monthly comparisons with percentages show groceries spending → Category-specific analysis Automatic daily summaries at your preferred time Database Integration All transactions are stored in PostgreSQL with proper schema: sql CREATE TABLE financial_transactions ( date DATE NOT NULL, category TEXT NOT NULL, type TEXT NOT NULL, amount NUMERIC(12,2) NOT NULL, person TEXT ); Setup Prerequisites n8n instance (self-hosted or n8n.cloud) WhatsApp Business Cloud API credentials PostgreSQL database (version 12+) OpenRouter API key for AI processing Quick Setup Steps Import the workflow template into your n8n instance Configure credentials: WhatsApp Business Cloud API (App Token + Phone Number ID) PostgreSQL connection details OpenRouter API key for AI processing Create database table using the provided SQL schema Test the connection by sending a sample message Customize the scheduled report timing (default: 8 AM daily) Verification Checklist [ ] WhatsApp webhook receives messages [ ] AI correctly parses transaction messages [ ] Database insertions work properly [ ] Confirmation messages are sent back [ ] Reports generate with accurate data How to customize this workflow to your needs AI Model Configuration Default: Uses OpenRouter with GPT-3.5-turbo for cost efficiency Upgrade: Switch to GPT-4 or Claude for better accuracy Local: Replace with self-hosted Ollama for complete privacy Database Options PostgreSQL: Recommended for production use Google Sheets: Alternative for simpler setups (nodes included) MySQL/SQLite: Easily adaptable with minor SQL modifications Message Classification Customize the classification system: 0: Reports (modify SQL queries for different analytics) 1: Transactions (adjust parsing rules for your language/currency) 2: Chat (customize AI responses for financial advice) Reporting Customization Scheduled reports: Change timing, format, and recipients Custom periods: Add quarterly, yearly, or custom date ranges Categories: Modify auto-categorization rules for your spending patterns Currency: Update formatting for your local currency Advanced Features Multi-user support: Add user identification for family/team use Receipt photos: Extend workflow to process image receipts via OCR Budgets: Add budget tracking and overspend alerts Integrations: Connect to banks via Plaid or other financial APIs Complete Package Included When you download this template, you get everything needed for immediate implementation: Ready-to-Use n8n Workflow Fully configured nodes with descriptive names explaining each step Color-coded sticky notes throughout the workflow explaining: What each branch does (Reports/Transactions/Chat) How the AI classification works Database connection requirements Error handling and troubleshooting tips Comprehensive Documentation Bundle Quick Start Guide: Get running in under 10 minutes Detailed Setup Guide: Complete configuration walkthrough with screenshots Branch Explanation Guide: Deep dive into each processing branch: Branch 0: Reports & Analytics - SQL queries and formatting Branch 1: Transaction Processing - AI parsing and database insertion Branch 2: Financial Chat - AI responses and conversation handling Built-in Workflow Documentation Sticky notes at every major step explaining the logic Node descriptions that clarify what each component does Visual flow indicators showing message routing paths Dependency callouts highlighting required credentials and connections Technical Implementation Details Database schema with complete SQL commands API configuration examples for all external services Troubleshooting checklist for common setup issues Performance optimization recommendations Bonus Resources Example message templates to test each workflow branch Sample data for testing reports and analytics Customization recipes for common modifications Integration patterns for extending functionality Example Usage Log Expenses: Spent 1200 on rent this month Paid 45 for gas at Shell Coffee 5.50 at Starbucks Log Income: Received 5000 salary from Company ABC Freelance payment 800 from Client XYZ Generate Reports: today's summary show this week's expenses compare March vs April spending how much on food this month? Expected Responses: ✅ Logged: expense | Rent | ₹1,200.00 | Landlord ✅ Logged: income | salary |₹12,000.00|company 📊 Today's Summary: Income: ₹0.00 Expenses: ₹1,245.50 Savings: -₹1,245.50 📈 March vs April: Expenses: ₹15,000 vs ₹12,500 (-16.7%) Top categories: Rent, Food, Transport
By Roshan RamaniMonitor security logs for failed login attempts with Slack alerts
How It Works: The 5-Node Anomaly Detection Flow This workflow efficiently processes logs to detect anomalies. Scheduled Check (Cron Node): This is the primary trigger. It schedules the workflow to run at a defined interval (e.g., every 15 minutes), ensuring logs are routinely scanned for suspicious activity. Fetch Logs (HTTP Request Node): This node is responsible for retrieving logs from an external source. It sends a request to your log API endpoint to get a batch of the most recent logs. Count Failed Logins (Code Node): This is the core of the detection logic. The JavaScript code filters the logs for a specific event ("login_failure"), counts the total, and identifies unique IPs involved. This information is then passed to the next node. Failed Logins > Threshold? (If Node): This node serves as the final filter. It checks if the number of failed logins exceeds a threshold you set (e.g., more than 5 attempts). If it does, the workflow is routed to the notification node; if not, the workflow ends safely. Send Anomaly Alert (Slack Node): This node sends an alert to your team if an anomaly is detected. The Slack message includes a summary of the anomaly, such as the number of failed attempts and the IPs involved, enabling a swift response. --- How to Set Up Implementing this essential log anomaly detector in your n8n instance is quick and straightforward. Prepare Your Credentials & API: Log API: Make sure you have an API endpoint or a way to get logs from your system (e.g., a server, CMS, or application). The logs should be in JSON format, and you'll need any necessary API keys or tokens. Slack Credential: Set up a Slack credential in n8n and get the Channel ID of your security alert channel (e.g., security-alerts). Import the Workflow JSON: Create a new workflow in n8n and choose "Import from JSON." Paste the JSON code (which was provided in a previous response). Configure the Nodes: Scheduled Check (Cron): Set the schedule according to your preference (e.g., every 15 minutes). Fetch Logs (HTTP Request): Update the URL and header/authentication to match your specific log API endpoint. Count Failed Logins (Code): Verify that the JavaScript code matches your log's JSON format. You may need to adjust log.event === 'login_failure' if your log events use a different name. Failed Logins > Threshold? (If): Adjust the threshold value (e.g., 5) based on your risk tolerance. Send Anomaly Alert (Slack): Select your Slack credential and enter the correct Channel ID. Test and Activate: Manual Test: Run the workflow manually to confirm it fetches logs and processes them correctly. You can temporarily lower the threshold to 0 to ensure the alert is triggered. Verify Output: Check your Slack channel to confirm that alerts are formatted and sent correctly. Activate: Once you're confident in its function, activate the workflow. n8n will now automatically monitor your logs on the schedule you set.
By Marth