What it is
Source: My own description of the project; Zerodha, NSE, Zerodha Varsity, the Corporate Finance Institute, Wikipedia and Anthropic for the terms. Date: 15 October 2026. Scope: Phase 1 and its paper test, as I report them. No live trading results or returns are claimed, and nothing on this page is financial advice.
Clautra is an options-selling agent for the Nifty 50 and Nifty Bank indices. It is powered by Claude, the AI model made by Anthropic, and it works through Zerodha’s API. Built
The Nifty 50 is an index of stocks listed on the National Stock Exchange of India, and Nifty Bank is made up of the most liquid Indian banking stocks. Options on both trade on the exchange (Nifty 50 F&O, Nifty Bank F&O).
What it does
- Options selling. The strategy the agent is built for, in which the seller of an option receives its premium (Zerodha Varsity).
- Data. A pipeline for the options chain and the Greeks, which measure how sensitive an option’s price is to factors such as volatility and the underlying’s price (Corporate Finance Institute). It also analyses open interest (OI), the number of outstanding contracts not yet settled (Wikipedia), and keeps an event calendar.
- Two-layer filter. A filter in two layers, with 8 quantitative and 6 technical checks. Reported
- Paper testing. Phase 1 was paper-tested, in simulated trading without real money (Corporate Finance Institute). The tests showed it blocking trades correctly in high-volatility conditions. Method
How it is built
Clautra is written in Python, powered by Claude, and works with Zerodha through Kite Connect, Zerodha’s API. Built It is built in phases, and Phase 1 covers the data and the filters.
Stack. Python, Claude, Zerodha Kite Connect
Where it stands
Phase 1 is complete. Built No figures from the paper test are shown here. No live trading results or returns are claimed, and nothing here is financial advice.
My part
I designed Clautra and built Phase 1. Built
