How GammaRips works
Better contracts every night. A real trade plan in the AI you already use.
Every night GammaRips ranks the US options market by liquidity and builds a pool of roughly 40 to 50 calls with deep books. Your AI reads the pool and the exit history over MCP and builds the trade plan with you: candidates, entry, target, stop, and size. You decide.
Better-quality contracts every night
The 100 most liquid optionable names, bullish only, one call per name chosen on contract liquidity. Deep books you can enter and exit near the quote.
The exit lab
How often past pool contracts hit +25%, +50%, and +100% within 3 trading days, and how often they touched a stop. Your AI sets the target and the stop from that history.
Your AI builds the trade plan
In Claude, ChatGPT, Cursor, or Codex. Candidates, entry, target, stop, and size, built for your account and your risk. You decide.
The pipeline, step by step
One rule, run once a night, in this order. Liquidity decides membership. Options flow rides along as context.
- About 3,500 optionable US stocks. The starting universe, refreshed weekly and scanned every night.
- Liquid names only. The stock traded 3M+ shares that session and its chain carries 25+ listed strikes.
- The top 100 by combined liquidity rank. The z-score of chain dollar volume plus the z-score of share volume.
- Bullish names only. Calls only.
- One out-of-the-money call per name, chosen on contract liquidity. The engine ranks the most liquid names, then selects the most tradeable contract inside each one.
- A published rank cuts to 50 most days. Most nights more than 50 bullish names qualify. A deterministic point-in-time rank (delta band, 60-day momentum, a liquidity demotion) keeps 50. It uses no outcome data and no learned weights.
The result is a pool of roughly 40 to 50 contracts, live by the open on /signals (free, no account) and in structured form on the MCP for your AI. This funnel has run since 2026-08-24.
Earnings: your AI checks each candidate
The published pool is not earnings-screened, so a name can report before its contract expires. The trade-plan playbook checks earnings before expiration for every candidate your AI shortlists, and drops the ones that fail.
Better-quality contracts, in numbers
Liquid contracts are the ones you can enter and exit near the quote. A thin contract costs you on the way in and again on the way out. Since the liquidity rule went live, the pool contract carries a much deeper book:
| Pool contract | Before the rule | Since Aug 24, 2026 |
|---|---|---|
| Median open interest | 906 | 4,705 |
| Median session volume | 232 | 1,069 |
| Open interest of the thinnest 10% | 29 | 1,606 |
| No print at the 10:00 ET entry (study) | 40.5% | 6.1% |
Since the rule: N = 1,725, pool contracts Aug 24 to Oct 15, 2026. Before: N = 3,501, the 60 scan days before the rule (May 27 to Aug 21, 2026). The no-print row is a 60-session study ending Aug 14, 2026 (N = 3,190 vs 545 contracts), not a live rate, counting a real print inside the 09:55 to 10:15 ET window. Quality here means execution: liquidity, open interest, volume, and fills.
Past results: real-money trades and historical pool data. Educational only. Not investment advice.
The exit lab
Pool contracts move a lot, in both directions. Of 793 expired pool contracts, scan dates Aug 24 to Oct 1, 2026, 54.6% hit +50% before expiration and 73.0% hit +20%. Within 3 trading days, 36.3% of 1,613 hit +50%, and 73.2% also touched −30%. Most contracts move both ways, so the exit plan your AI sets decides the result. The exit lab gives your AI the history to set the target and the stop on purpose, instead of a guess.
Touch rates
How often past pool contracts hit each profit level and each stop, with your own filters.
The opportunity surface
For every past pool contract, how far the premium moved in its favor and against it over the 3 trading days after the 10:00 ET entry, with no exit applied.
Bracket scoring
Your AI proposes a target, a stop, and a hold. The lab scores that bracket on past pool contracts.
Contract replay
The minute or daily price tape of any past pool contract, so your AI can see how a move actually unfolded.
A hit is a touch: the premium traded at that level inside the window. The full hit rates are on the Track Record, next to the live agent record (11 of 14 real-money trades closed at a profit, +$2,263 net, Sep 14 to Oct 2, 2026).
Past results: real-money trades and historical pool data. Educational only. Not investment advice. Pool figures are tracked on a paper basis.
Your AI builds the trade plan
Ask your AI for a trade. With GammaRips connected, it follows the start-here playbook on the MCP:
- Checks the session and the pool. Is the market open, and is tonight’s pool fresh? If the market is closed, your AI plans for the next session.
- Checks the regime. The regime rail passes when VIX is at or below VIX3M. Your AI reads it before it plans a trade.
- Reads the pool and shortlists 1 to 3 names. Each name comes with its thesis, technicals, catalyst, and selected contract. Your AI filters on your own criteria.
- Checks each candidate. Earnings before expiration, and fresh open interest and volume on the exact contract. A name that fails drops out.
- Plans the exit from history. The exit lab gives the touch rates for any target and stop, the typical favorable and adverse moves, and a score for the bracket.
- Gives you the plan. The contract, why it passed, entry near 10:00 ET with a limit order near the mid, target, stop, maximum loss, and a size that keeps that loss inside your risk budget. Or a reasoned no-trade, with the check that failed.
Why not one shared trade for everyone? Shared trades get crowded. A plan your AI builds from the data, for your account and your risk, does not. That is why the MCP serves the pool and the history, and your AI does the planning.
The nightly clock
23:00 ET: the scanner walks the universe, prices full option chains across the liquid names, and ranks them. Overnight: the pool is enriched with news context, technical levels, flow dollars, and one contract per name. By the open: the pool is live on /signals and on the MCP.
Around 10:00 ET: the reference entry time. The exit lab measures every past contract from this mark, and your AI can check fresh open interest and volume on any contract before it enters.
What is unusual options activity?
Unusual options activity (UOA) is options volume that runs far above normal levels for a particular stock. It can mean institutional traders, such as hedge funds, pension funds, or large trading desks, are building new positions.
The usual indicators are the volume-to-open-interest ratio (fresh activity against positions already on the books), dollar flow (how much capital moved), and directional imbalance (calls against puts).
How GammaRips uses it. Since 2026-08-24, membership in the pool is liquidity-based. Flow does not decide who gets in. We still measure it and publish it, because it is useful context for an AI reading a name. It is a column in the data, not a gate in the funnel.
The flow score, in plain English
Each name in the pool carries a 0 to 10 flow score. It is a checklist, not a model. The scanner asks a few plain questions about where option money went:
- One-sided money. Option dollars piled onto one side, calls well over puts.
- New money. The day's trading against the positions already on the books. Fresh bets, not old ones adjusting.
- Built like an institution. Buying spread across several strikes, not one lottery ticket.
- Size of the new positioning. How many dollars landed on that side.
- The stock moved too. Whether the price confirmed with a real move on the day.
Small bonuses when the money bets against the tape, such as heavy call buying on a red day, or when a whole industry lights up the same direction at once.
Reading it: the score counts evidence of positioning. It is context for your AI, not a forecast, and it does not rank the pool or decide membership. Its floor of 1 is cosmetic: nearly every liquid name clears it.
What is agentic trading, and how do you try it?
Agentic trading means your AI (Claude, ChatGPT, Cursor, Codex, or one you build) works as your own options analyst. You give it real data. It reads tonight's pool, checks how similar contracts moved, sets an exit from that history, and lays out a plan. The decision, the sizing, and the trade stay yours.
A chatbot without data improvises. Its knowledge froze months ago, and no options-flow data exists in any training set. The fix is a connected model, not a smarter one.
Here is the on-ramp. Free, no account: browse tonight's pool and the Track Record. Free, no card: connect any MCP client to the free tier for the pool preview, the daily report, and the playbooks. Agent Access: the full pool, liquidity checks, and the exit lab, so your AI can build the whole trade plan. Setup takes minutes on the For Your Agent page.
Your analysis, your account
GammaRips publishes options-flow data, a real-money agent record, and educational content. The pool is the output of a mechanical engine, not personalized advice, and anything your AI concludes from the data is your analysis. You trade your own account. GammaRips does not manage your money. Past results do not promise future results. Educational only. Not investment advice.
Put your AI on tomorrow's pool
Install GammaRips in the AI you already use, then ask it for a trade plan.