athenara:~$ registry show agents/thetagang
ThetaGang
An Interactive Brokers bot that started as an implementation of the options Wheel strategy and grew into configurable portfolio automation with share rebalancing and hedging.
added 2026-08-17 · AGPL-3.0-only · external
$ pip install thetagang $ curl -Lq https://raw.githubusercontent.com/brndnmtthws/thetagang/main/thetagang.toml -o ./thetagang.toml $ thetagang --config ./thetagang.toml --dry-run
requires python >=3.10,<3.15
you also need: Interactive Brokers account, IBC / IB Gateway
ThetaGang connects to an Interactive Brokers account and submits and rolls real option and share orders from a rules-based TOML configuration. It began as a basic implementation of “The Wheel” — selling cash-secured puts, then covered calls against assigned shares — and has since grown into a broader portfolio automation tool covering direct share rebalancing, cash management, VIX call hedging, long-put tail hedging, regime-aware rebalancing and exchange-hours gating. The shipped default is a diversified portfolio of SPY (40%), QQQ (30%), TLT (20%) and smaller individual positions. Run state — order activity, executions, historical bars, account snapshots and decision gates — can be persisted to SQLite and reused across runs.
The strategy rests on an assumption the README states outright: “this strategy is based on the assumption that implied volatility is, on average, always higher than realized volatility. In cases where this is not true, this strategy will cause you to lose money.” No backtested or live returns are published; what the project publishes instead is a warning that options trading involves substantial risk and that selling naked puts carries theoretically unlimited downside. It needs a funded IBKR account plus IBC or IB Gateway, and the README recommends starting on an IBKR paper account. Because its decisions are account-level, running materially different strategies in the same account can produce harmful interactions.
The license is AGPL-3.0-only — network copyleft, which matters to anyone considering it as the
basis of a hosted service. Distribution is via PyPI and a published Docker image; --dry-run
previews the orders it would place without submitting them.
trading [●●●●·] advanced ai [●····] none programming [●●···] basic setup [●●●●·] advanced
athenara:~$