AUTONOMOUS · LONG-ONLY · THE EQUITIES BOOK: S&P 500 · NASDAQ-100 · MACRO
Coil is the discipline, not the trader. A decade of tested rules — rank the market's leaders, buy them coiled at real support, size by conviction, exit by rule, hold cash when nothing qualifies — packaged so your own AI agent runs them out of the box. Scheduled Claude Code sessions cover every name in the Equities book (S&P 500, Nasdaq-100, and Macro — ~560 names), on your machine, through your brokerage, with your keys. It ships with live trading OFF, and one command pulls the plug.
The performance figures below are a survivorship-free research backtest that validates the ranking brain — not the live account, not typical, not a prediction. The engine is newly live. It uses leveraged ETFs, which can move fast and lose money, including total loss.
Launch price — $29 until the closing bell, . $49 after.
Just want the scored board? Coil Scanner — $12/mo, nothing to install →
Building an agent? Free delayed MCP server, free API key, pay-per-read x402 →
You set it up once, read the rules for yourself, and arm it. After that it runs hands-off on your own account.
Drop the packet into your agent. It becomes the operator that wakes on a schedule, scans the whole market, reads the scores, and works the leaders for you — placing trades and managing exits — so you don't have to sit and watch. It executes the rules; it can't rewrite them.
Link your account through a broker connector and add a free data key. Your credentials stay on your machine — nothing phones home, and you can revoke access any time.
No Robinhood yet? Open an account with my referral ↗ (referral)
Once armed, it takes entries, manages exits by rule, and enforces hard safety limits that cut size or stand down when losses pile up. Each night it retunes position sizing — only sizing — to your own fills, within strict, reversible limits. It tunes position sizing to your tape; it never rewrites the strategy.
Setup runs about an hour to a scheduled dry-run: point your agent at the folder, add your keys, run the self-test suite (100+ assertions), watch it work. It ships disarmed — going live is a separate, deliberately gated step — you read every rule, and one command shuts it all down.
The framework refined over years of live trading, encoded so it runs the same whether you're watching or not. Tap "under the hood" on any card for the exact rules.
Coil scores every name in the Equities book, then buys only the highest-conviction leaders that are ready — coiled at real support, not falling and not overbought — and sizes them by conviction. It ignores everything else.
The moment a position stops working — the leg statistically ends, or the trailing stop is hit — Coil is out, by rule; it doesn't wait for a big loss to admit a trade failed. It scales out on a ladder into strength, and every position carries a structural stop from the moment it opens.
An immutable safety core clamps everything: account lockdown, position caps, drawdown circuit breakers, kill switches — written to disk so they survive restarts. They reduce big single-day damage; they don't remove risk.
Coil doesn't marry a sector. It scores the whole market and follows leadership wherever it goes — semis led in 2024, something else led before and after — buying only names coiled at real support, never the falling knives or the overbought chasers.
In every market some group is doing the heavy lifting. Miss the rotation and you're early, late, or in last cycle's winners. Coil scores every name in the Equities book, so it's in the leaders now.
When semis led, Coil's book filled with them. When leadership rotated, so did Coil. The scanner re-ranks the whole market every cycle — sector-rotation phase and market posture included.
It buys weakness inside strength — leaders coiled at support that are ready — and stands aside from the falling knives and the overbought chasers. The setup is the same everywhere; only the names change.
Leverage cuts both ways: the leveraged ETFs Coil uses to accelerate its best ideas decay over time and can lose money fast, including large or total loss.
The hardest part of trading isn't finding a good name — it's doing it the same way every time, without fear or FOMO. Coil scans the whole market and follows its written rules whether you're at your desk, asleep, or out living your life. Same rules, no flinching — that consistency is the point, not a shield against losses.
Same market, same years — the only difference is whether emotion's in the loop. Coil runs the rules-based side: it doesn't chase the euphoria or panic-sell the bottom. Watch what that does over time.
Hypothetical, illustrative traders starting at $100,000 — not Coil's actual or expected returns, not a track record, and not a prediction. Coil's real research-validated figures are in the Proof section below. Trading leveraged ETFs can lose money rapidly, including total loss.
This is the leadership-rotation research backbone the whole system is built on — point-in-time S&P 500 membership including delisted names (FRC and SIVB are in the data, so no survivorship bias), next-open fills, and trading costs, from 2017 through 2026 H1. Each year is the Coil ranking brain against buy-and-hold SPY. Read it straight: in calm bull years SPY often wins, and Coil pulls ahead in the bear and in the leadership regimes.
Read it honestly: through end-2025 the ranking brain ran about even with SPY at roughly a third less drawdown — the outperformance concentrates in leadership regimes (2025 +51%, 2026 H1 +86%). We publish this because you should know when it wins and when it doesn't.
Backtests are research, not a promise of returns. These figures validate the ranking brain Coil's scanner is built on — they are not a record of live account returns. The live engine is newly running and trades long-only across the Equities book (S&P 500, Nasdaq-100, and Macro) with risk-based sizing and leveraged-ETF acceleration; live results will differ, in both directions.
Basis: point-in-time S&P 500 membership including delisted names (survivorship-free), next-open fills, modeled trading costs, 2017 → 2026 H1, compared to buy-and-hold SPY over the same window. Every figure is re-runnable yourself with the included harness.
How it scores leaders, picks entries, and exits
When the pool has no ready leaders — or the market posture turns defensive — Coil holds cash instead of forcing a trade. On Robinhood Gold, that uninvested cash earns the broker's cash sweep — ~3.35% APY (variable) as of Feb 2026. That yield is the broker's, not Coil's.
~3.35% APY is variable, set by Robinhood, and requires Robinhood Gold (a paid tier) — the broker's yield on idle cash, not Coil's, not risk-free.
New to Robinhood? Open an account with my referral link ↗ (referral link — I may receive a bonus)
Figures are a survivorship-free research backtest across 2017 → 2026 H1, compared to buy-and-hold SPY. They are not a record of client or account returns and not a promise of future performance. Trading leveraged ETFs can lose money rapidly, including total loss.
Anyone can fit a backtest that looks perfect — cherry-pick the window, quietly drop the companies that went bankrupt, draw the smooth hockey stick. We did the opposite: kept the delisted failures in the data, showed every year the ranking brain trailed SPY right on the grid above, and published the drawdown next to it. A number you can't trust is worse than no number.
And we'll say it plainly: the v4 engine is early — newly live, with no client track record yet, and the ranking brain is validated on research, not a live account. So we hand you the validation harness to re-run every figure yourself, rather than ask you to take ours on faith.
The same scanner brain, two ways to use it. Coil Scanner puts the daily scored board in your browser — read it in two minutes each morning. The Coil Trading Bot is the whole system in one download: the same scanner running locally, the dashboard, and the agentic trading engine your own AI agent operates.
The same scored board the Trading Bot trades from, published to your browser each market morning. Nothing to install, no brokerage connected, no order ever placed. You make the calls.
An impersonal research publication — scores and states, no stop/target prices, not advice.
Why $49 is crossed out: $29 is the launch price and
it ends at the closing bell on .
After that it is $49 and stays there.
The same scored board, running on your machine, plus the engine that turns it into orders. Your own AI agent places them at your brokerage — scan, decide, place, manage, by rule. One download, yours forever.
Live trading ships OFF — you arm it deliberately. Requires Claude Code and your own brokerage account. All sales final — instant digital download, yours to keep and run forever.
Same brain, same scores. The only difference is whose hands are on the order button. With the Scanner, yours. With the Trading Bot, your own AI agent's — it needs Claude Code you install, a brokerage account you own, and live trading stays off until you deliberately arm it.
Every copy tunes itself to your own trading. At each close a local pass reads your own closed round trips and leans position sizing toward what has been working — by book (S&P 500 / Nasdaq-100 / Macro) and by buy-score tier. It's size-only: it never touches which names it buys, the entries, the exits, the stops, or the caps.
The tilts are hard-clamped (about 0.55×–1.35× overall) and shrinkage-weighted, so one trade barely moves the dial, and it stays neutral until you've armed the engine and closed a couple of real trades.
It runs entirely on your machine — nothing phones home, nothing leaves it, and your history persists across packet updates — with every tilt written to a readable audit trail. New validated versions — new scanner factors, entry windows, or book coverage — are cold-validated first and announced on coil.trade; you choose if and when to grab them.
You own the code. Your agent runs it. Your broker, your keys.
The Coil engine is software you operate yourself; Coil Board is an impersonal research publication. Neither is a managed account, personalized advice, or a promise of profit.
Coil is built to be discovered, installed, run — and now bought — by an AI agent. An agent with a funded wallet can purchase a read of the live scanner board directly, by itself: $0.25 in USDC on Base over the x402 payment protocol — no account, no card, no email, settlement receipt on-chain.
For the products humans buy — the $29 Trading Bot, the $12/mo Scanner — checkout is a hosted Gumroad flow (Gumroad is the merchant of record: checkout, receipts, license keys), and an agent with delegated card authority can complete it like any web checkout, with you in control. Discovery is machine-readable end to end: /llms.txt, /.well-known/coil.json, an OpenAPI contract at /openapi.json, and listings in the x402 agent directories.
Non-technical? You've already got the full picture above. Curious how it works? Read on for the compression-to-ignition entry logic, the exit math, the self-improvement loops, and the validation harness.
A sharp expansion in a market leader isn't a trend you chase — it's born as a short-timeframe reversal inside the higher-timeframe trend. That's the coil, and it's the same setup in every sector. Coil buys it long-only: disciplined weakness inside strength — an oversold flush, a pullback to trend, a momentum reclaim — in the names the scanner ranks as leaders and ready. It waits for the compression, rides the ignition, and gets out the instant the move statistically ends. No hope, no targets, no averaging down.
The signals — leadership, the entry window, momentum, the sector-rotation phase, and the market posture — fire through the engine and resolve to one verdict: long a leader (and accelerate it), rotate risk-off toward the Macro book, exit, or stand down to cash. It never shorts. Every position carries a structural stop, and the immutable safety core clamps everything. The same wiring every time, with no mood and no FOMO.
See the full method — compression-to-ignition entries, the exit math, the validation harness
A bounded local pass leans position sizing toward what has been working on your own closed trades — size-only, by book and buy-score tier, hard-clamped 0.55×–1.35× and shrinkage-weighted. Hash-pinned safety checks guard the frozen, validated ranking brain, which never mutates on your machine. New structure arrives only as a deliberately shipped, re-validated version.
The three ways a backtest lies are look-ahead bias, survivorship bias, and next-bar leakage — each one flatters the numbers. Coil's research backbone is built against all three: point-in-time membership that keeps the delisted failures (FRC, SIVB) in the data, next-open fills, and integrity checks for look-ahead. A self-test suite of 100+ assertions — safety spine, scanner factors, the bounded sizing loop, indicators — now stands watch, and you can run it yourself.
That's the whole pitch. A system that audits itself harder than you would is the only kind worth running unattended.
An autonomous, long-only trading system you run yourself, in three parts and one download. The scanner scores every name in the Equities book (S&P 500, Nasdaq-100, and a Macro book of bonds, income, metals, commodities — ~560 names) — opportunity, entry-window, hold-conviction, leadership, sector-rotation phase, market posture — and maps leaders to leveraged vehicles (NVDA → NVDL, QQQ → TQQQ). The dashboard is the command view. The engine trades the scanner's published signals: it buys the highest-conviction leaders when they're ready, sizes by conviction, accelerates with leveraged ETFs, exits on a ladder with trailing stops, and holds cash when the pool is thin — all behind an immutable safety core. It runs on your machine, with your agent, your broker, and your keys.
It's built for Claude Code — scheduled Claude sessions are the operator — and adaptable to your agent of choice. It places trades through a broker connector, built for Robinhood, and works with any broker that has an equivalent MCP. Under the hood it's Python on Mac or Windows (the agent-led onboarding detects your OS) with a free Alpaca data key. You add your own broker account id and key, and your credentials never leave your machine.
Once armed, it runs hands-off: it scans the market, takes the entries, manages every exit by rule, and applies risk circuit-breakers tied to your account's high-water mark — no screen-watching required. You're never locked out: read every rule, see exactly what it did and why on the local dashboard, and stop it in one command. The circuit-breakers fire by rule, without emotion — they reduce big single-day damage, they don't remove risk. It uses leveraged ETFs to accelerate its best ideas, and those can move fast and lose value rapidly, including total loss.
All three parts in one download: the market-wide scanner (opportunity, entry-window, hold-conviction, leadership, sector rotation, market posture, with leveraged-vehicle mapping), the command dashboard (books, scorecards with why-this-score notes, top picks, structural stops and targets), and the long-only autonomous engine (conviction-ranked entries, leveraged acceleration, ladder exits, trailing stops, cash when thin). Plus the immutable safety core with persistent circuit-breakers, the local sizing-only learning loop, the validation harness with a self-test suite of 100+ assertions you can run yourself, and the plain-English onboarding. No credentials are included — you add your own.
Because leadership rotates. Every cycle has leaders — semiconductors led in 2024, but before and after that the baton passed to other groups. A one-sector engine is right only while its sector leads and dead weight the rest of the time. Coil scores the whole market — the Equities book of S&P 500, Nasdaq-100, and Macro names — and follows leadership wherever it goes, buying names coiled at real support when they're ready and standing aside from the falling knives and the overbought chasers. The name still fits: Coil looks for leaders coiled at support and strikes when they're ready; it just does it across the whole market now.
Very. Coil accelerates its highest-conviction ideas with 2× and 3× leveraged ETFs, so a ~10% move in the underlying is roughly a ~20–30% move in the vehicle, before gaps and slippage. Leveraged ETFs also decay on multi-day holds because they track daily moves, and a position can gap through the hard stop before the open (a stop can't fill in a gap). The engine caps exposure and the circuit-breaker ladder reduces single-day damage, but neither removes risk. You can lose money rapidly, including total loss.
The engine only buys when the pool has real leaders that are ready — coiled at support, not falling and not overbought. When the market posture turns defensive it tightens entries and rotates toward the Macro book (bonds, income, metals); when nothing qualifies it holds cash, and on Robinhood Gold that idle cash earns the broker's cash sweep, ~3.35% APY (variable) as of Feb 2026 — the broker's yield, not Coil's, and not guaranteed or risk-free. The research backbone is honest about when the edge shows up: through end-2025 the ranking brain ran about even with SPY at roughly a third less drawdown, and it pulled ahead in leadership regimes (2025 and 2026). Those are research figures, not a prediction.
The engine, no — the $29 download is a long-only equities engine and does not trade crypto. The board, yes as a read: since July 15, 2026 the Coil Scanner board pairs the Equities book with the Crypto book — BTC and ETH scored with long-only trend signals. The honesty rider ships with it, ugly parts first: the cold backtest does not beat buy-and-hold CAGR (59.6% vs 74.9% for a 50/50 BTC/ETH hold, 2016–2026); what it buys is smaller crashes (2022: 0.0% vs −66.0%; max drawdown −64.3% vs −87.4%) at the cost of whipsaw years (2024: −22.6% vs +74.7%). Signals tracked live since July 15 — backtest before that, not traded with real money. The full numbers, losses included.
You buy the current version once and keep it. New validated versions — new scanner factors, entry windows, exits, or book coverage — are cold-validated first and announced on coil.trade; there's no auto-update service and no subscription. What you download runs forever, offline, on your own machine, and its local loop keeps tuning position sizing to your own fills within hard bounds.
Running it is what Coil is built for: an agent can discover it (there's a machine-readable manifest at /.well-known/coil.json), install it, and operate it on a schedule. Buying is ordinary web commerce: checkout is a hosted Gumroad flow — Gumroad is the merchant of record and handles receipts and license keys. It's a single one-time purchase, and all sales are final (instant digital download). There's no special agent-checkout integration; an agent with delegated card authority can complete the hosted checkout like any web checkout. And the scanner board itself is agent-purchasable natively: $0.25 per read in USDC over the x402 protocol — no account, no card, no email (the agent endpoint). It runs on your approval and your own credentials, and you stay in control. To be clear: this is software you run yourself. It's not investment advice and makes no income promise.
About an hour to a scheduled dry-run. You point your agent at the folder and the interactive onboarding takes over: it detects your OS (Mac or Windows), sets up a free Alpaca data key, and runs first light — the agent introduces itself, scans the market, and opens the dashboard with live scores within minutes. Then it watches in dry-run. Going live is a separate, deliberately gated flow: your own account allowlisted, an integrity re-pin, the self-test suite green, and a typed total-loss acknowledgment. It ships disarmed.
The ranking brain — the leadership scoring the whole system is built on — was replayed on a survivorship-free research backbone: point-in-time S&P 500 membership including delisted names (so failures like FRC and SIVB are in the data, not quietly dropped), next-open fills, and modeled trading costs, from 2017 through 2026 H1. That's research that validates the brain, not a live account record, and it has self-tuning degrees of freedom, so treat the figures as evidence, not proof. The live engine is newly running; live and forward results, accumulated on small size over time, matter more than any backtest — and every figure is re-runnable yourself with the included harness.
Coil trades actively — many entries and exits a year across a rotating set of names — so realized gains are largely short-term (taxed as ordinary income), and active trading can create wash-sale complications that defer losses. None of that is tax advice — talk to a tax professional about your own situation before running it with real money. Coil is software you operate yourself and reports nothing on your behalf.
No. Coil is software you run yourself, with your own credentials and your own capital. It is not investment advice, not a managed fund, and not a guarantee of profit. Coil holds no funds and executes nothing on anyone's behalf. Trading leveraged ETFs carries substantial risk of loss, including total loss. The performance figures on this page are a survivorship-free research backtest that validates the ranking brain — not a record of anyone's client or account returns, and not a prediction.
Coil was built by Joey Fife — a product manager who trades for a living, running his own account since 2014. He didn't set out to sell software. He set out to get his time back. Trading well is a discipline tax you pay all day, every day — watch the tape, wait for the real entry, honor the stop, don't chase — and that tax was coming out of the hours that matter most: the ones with his family.
So he handed the discipline to an agent. Coil watches the market so he doesn't have to, buys only when a leader is actually ready, and manages every exit without emotion — validated on a survivorship-free decade of market history, with every failed idea logged alongside the numbers that killed it, and the years it trails the market shown, not hidden.
Discipline, autonomized.
He's not a financial advisor, and Coil isn't advice — it's the tool he runs himself, shared the way he'd want it shown to him.
The scanner, the dashboard, and the engine on your machine in about an hour — your agent, your account, your keys, your control.