Start to finish, in the order that actually matters. The first two steps depend on your broker and can take a day or more; everything after them takes minutes. Do them first.
Nothing here is a recommendation about how to trade. The app ships with no strategy and this page does not supply one. It tells you how to make the software work.
In the Webull app: Settings → API Management. Create an app key and app secret, and request both:
You need the live keys even if you only ever paper trade. That is not a typo in the instructions: market‑data entitlement attaches to the live key, so quotes come from the live endpoint while orders route to paper. With paper keys alone the app connects, shows your balance, and cannot price a single contract — which looks like a working app that never trades.
Request the paper keys first if you are asked to choose.
On the same account, subscribe to real‑time options market data (OPRA) under a non‑display or API agreement.
Non‑display is the one that matters. The ordinary real‑time subscription covers data a human looks at on a screen; data consumed by a program is a separate agreement. Ask for it by name. Get the wrong tier and your keys authenticate, the app connects, and your quotes are still delayed with nothing to tell you.
This is the one prerequisite with no software answer. A short‑dated contract priced between ten cents and a dollar moves a large fraction of its premium in seconds; on delayed data the app is reading the past and pricing orders against quotes that no longer exist. Expect approval to take a day or more, which is why it is step two and not step six.
More detail: what an options bot needs from Webull.
Get it from the download page. Windows is an installer; macOS is a disk image for Apple Silicon. On Windows the installer can fetch Google Chrome for you if you do not have it — the research step drives a real browser window, so it is required.
Your settings, trade history and credentials are never overwritten by a reinstall, so updating later is safe.
Paste all four keys into the Credentials panel and press Save credentials, then Login.
They go into your operating system's own credential store — Windows Credential Manager or the macOS Keychain — and never into a file. The account line then shows your mode, account id and buying power, which is how you know the connection is real rather than merely accepted.
Leave the mode on PAPER. Switching to LIVE makes you type the word LIVE to confirm, and is refused while armed.
This is the step people do not expect. Options Sniper ships with no
strategy at all. Every strategy value in the shipped configuration is
empty, and the engine refuses to read one that has not been set rather than
reaching for a default. FIRE stays disabled until a strategy
exists.
That is deliberate. A default is a recommendation, and a recommendation that ships to everybody is one nobody chose. If a number decides what you risk, you should have picked it.
So: click LOADOUT at the top right, then the PACK BUILDER tab. You get all fifty‑one parameters, grouped:
Every field states its type and its permitted range, and the ones that are rarely touched are behind the Advanced toggle, so the first pass is shorter than fifty‑one boxes sounds. The values the form starts with are a neutral starting point to edit, not a suggested strategy.
When you are done press Save pack. It writes a
.ntlp file wherever you choose — keep it somewhere you will
find again, like your Documents folder. You can build several and keep them.
Sizing deserves a moment. Buying power share is the lever that decides whether a losing streak is a drawdown or the end of the account, and it does far more damage than a target set slightly wrong. The simulator included with the app exists to compare settings against each other — it runs on a synthetic tick path, so treat it as a comparison instrument and not as evidence of an edge.
Still in LOADOUT, on the LOADOUT tab,
click Load and pick your .ntlp file. The card
fills in and FIRE becomes available.
Press FIRE. It tells you the mode and the next research time and asks you to confirm. After that it is armed and standing by.
Leave it running. Minimise it if you like — it is built to sit in the tray all day. If it is closed, nothing happens: no research, no trade. It does record that it was armed, and on its next launch it reconnects, reloads your pack and re-arms itself, saying so in the activity log. If any of that fails it says NOT ARMED rather than letting you believe otherwise.
JAM disarms it and closes an open position, mid‑trade if need be, cancelling the working bracket orders first so they cannot fight the closing order. It stops the strategy, not the application.
Trading options involves substantial risk of loss and is not appropriate for everyone. Nothing on this page is advice, no return is promised, and past results do not predict future results. Paper trade first.