Detect NSE options anomalies with Supabase, OpenAI, Google Sheets and Gmail

Quick overview

This workflow runs every 30 minutes to scan an NSE options watchlist from Google Sheets, fetch option chain data via HTTP, compare it with historical snapshots in Supabase, generate an OpenAI interpretation for anomalies, log results to Google Sheets, and send a Gmail digest alert.

How it works

1. Runs every 30 minutes on a schedule.
2. Reads a watchlist from Google Sheets and keeps only symbols marked as enabled, along with their scan settings.
3. Requests the latest option chain for each symbol from the configured HTTP endpoint and normalizes the returned contracts into a consistent structure.
4. Pulls historical option snapshot data for each symbol from Supabase and calculates volume ratio, open-interest ratio, and a z-score to flag anomalous contracts.
5. For anomalous contracts, saves the full option snapshot to Supabase and asks OpenAI to generate a concise plain-text interpretation with observation, possible interpretation, and a risk note.
6. Stores the AI-enriched anomaly record in a Supabase anomalies table, appends the anomaly to a Google Sheets log, and builds a single text digest.
7. Sends the digest email through Gmail to the configured recipient.

Setup

1. Connect Google Sheets credentials and update the spreadsheet ID and sheet tabs for the watchlist and anomaly log.
2. Configure the HTTP request URL to point to your NSE option chain data source and ensure it returns the expected contract fields (symbol, expiry, strike, option_type, ltp, volume, open_interest, change_in_oi, iv, underlying_price, captured_at, source).
3. Connect Supabase credentials and create/configure the required tables (option_snapshots and option_anomalies) with columns matching the fields the workflow writes.
4. Add an OpenAI API credential (and adjust the model, if needed) for generating the anomaly interpretation.
5. Connect a Gmail OAuth credential and set the recipient email address in the email-sending step.

Additional info

Use Case Examples

1. NSE Options Activity Monitoring

Automatically monitor configured NSE symbols and identify contracts showing unusual volume, open interest, or statistical activity.

2. Unusual Volume Detection

Identify contracts where current volume is significantly higher than the configured historical baseline.

3. Open Interest Monitoring

Identify option contracts where current open interest exceeds the configured anomaly threshold.

4. Statistical Anomaly Monitoring

Use the calculated z score to identify option contracts whose activity differs significantly from the calculated historical pattern.

5. Automated Market Activity Reporting

Store detected anomalies in Supabase, maintain a Google Sheets log, generate AI explanations, and receive a consolidated Gmail report.

There can be many more use cases based on the available market data and the additional calculations or notification channels added to the workflow.

Troubleshooting Guide

Issue	Possible Cause	Solution
No symbols are processed	Watchlist entries are disabled	Check the enabled column and set the required symbols to true
Watchlist data is not loading	Google Sheets credentials or sheet configuration is incorrect	Verify the Google Sheets credential, document, worksheet, and column names
HTTP Request returns no contracts	The configured endpoint is unavailable or its response structure changed	Test the endpoint and verify that it returns the expected data collection
Contract fields are empty	The HTTP response does not match the expected structure	Compare the response fields with the fields required by Normalize Option Contracts
Historical data is missing	Supabase table has insufficient records	Verify that option_snapshots contains historical records for the scanned symbol
No anomalies are detected	Current values do not meet the configured thresholds	Review the volume ratio, OI ratio, and z score calculations
Too many anomalies are detected	Thresholds are too low	Increase the anomaly thresholds in Calculate Option Anomaly
OpenAI interpretation is missing	OpenAI credentials or response mapping may be incorrect	Verify the OpenAI credential, model configuration, and output field
Supabase anomaly record is not created	Supabase table or field mapping is incorrect	Verify the option_anomalies table and mapped fields
Google Sheets log is empty	Spreadsheet mapping is incorrect	Verify the anomaly log worksheet and column names
Multiple emails are received	The email node is receiving multiple items	Ensure Build Email Alert Digest combines the incoming items into one output item
Gmail email contains no anomalies	Email preparation fields are empty	Check the output of Build Email Alert Digest before executing Gmail
AI interpretation is not saved	The OpenAI response field does not match the configured mapping	Check the OpenAI node output and update the ai_interpretation mapping
Workflow does not run automatically	Workflow is inactive or schedule configuration is incorrect	Activate the workflow and verify the schedule configuration

Need Help

Setting up an options monitoring workflow requires careful configuration of market data, historical calculations, database storage, AI processing and notification logic.

If you need help configuring this workflow, modifying anomaly detection rules, connecting a different market data source, improving the OpenAI analysis, creating dashboards, adding notification channels or building a similar n8n automation for your business process, the WeblineIndia team can help with setup, customization, integration and ongoing workflow improvements.