Stress-test personal finances for life events with Plaid-style data and Claude

Quick overview

This workflow collects a user-selected financial life event through an n8n Form, retrieves account and market data, runs a Monte Carlo stress test, and uses Anthropic Claude to generate and email a personalized financial report.

How it works

1. Receives a form submission that specifies the life event to test (buy a house, lose a job, or both) and the simulation horizon and parameters.
2. Fetches linked account balances and recent transactions from a Plaid-style account-aggregator API and fetches historical market return statistics from a market-data API.
3. Calculates a financial baseline (net worth, estimated monthly income/expenses, and emergency-fund coverage) and packages the life-event and market inputs for simulation.
4. Runs 1,000 Monte Carlo month-by-month simulations applying the selected event’s cash-flow shocks and random market returns, then summarizes percentiles and shortfall probability.
5. Sends the simulation results to Anthropic Claude to produce a JSON risk narrative, risk level, and three recommendations, then formats a complete plain-text report.
6. Sends the digital twin report by SMTP email with the computed risk level in the subject line.

Setup

1. Add Anthropic credentials and select the Claude model in the Anthropic Chat Model node used by the AI agent.
2. Configure the account-aggregator HTTP request with your Plaid (or equivalent) client ID, secret, and access token, and adjust the endpoint/response mapping if your provider differs.
3. Add your market-data API key in the Authorization header and update the market-data endpoint to return mean and standard deviation of monthly returns.
4. Set up SMTP credentials and replace the from/to email addresses used to send the final report.
5. Activate the workflow and use the n8n Form URL to submit an event scenario and run a simulation.