From running your first AI analysis to reading the output, macro & news, sniper entries, journaling and billing — the complete guide.
What AthenixFlow is, how to get in, and how the platform is organized.
AthenixFlow is a Smart Money Concepts market-analysis terminal for Forex, Crypto, Stocks and Metals — every read is scored on the server and can be graded against real price. You choose a trading mode, an instrument and a timeframe, and the engine returns a structured read of the market — bias, liquidity, points of interest, scenario weights, and a proposed trade setup (entry, stop, and take-profits) — in the language of Smart Money Concepts (ICT/SMC).
After you log in, the Dashboard is your home base. It shows your subscription tier, your two credit balances (Analysis and Education units), your live terminal/database status, a Trade Performance snapshot (your TP1 hit rate and outcome breakdown from the analyses you’ve logged results on), and Quick-Launch cards that jump straight to the AI Assistant, Education Hub, Trade Setups and Trade Journal.
The left sidebar is your main navigation: Dashboard, AI Assistant, Education Hub, Trade Setups, Trade Journal, Partners, Subscription, Settings, and this Documentation page. On mobile, tap the menu icon to open the sidebar drawer.
The platform runs on two credit types: Analysis units (one is spent per market analysis) and Education units (one is spent per mentor question). Units are added to your balance whenever you buy — a Starter Pack, a top-up or a pass — and they never expire. See “Passes, Units & Billing” below for details.
The Smart Money / ICT vocabulary the engine uses — read this first if the terms are new.
AthenixFlow speaks in Smart Money Concepts (SMC), the framework popularized by ICT. The engine reads markets as a story of liquidity: price is engineered to reach pools of resting orders, then reverses or expands. These are the terms you will see throughout your analyses.
How to configure and run an analysis in the AI Assistant.
The AI Assistant is the core of AthenixFlow. The left column is where you configure the analysis; the right column shows the result and your history.
The mode sets the timeframe band and how far down the engine refines the entry. Refinement steps down progressively — H4 to H2 to H1 to M30 to M15, for example — and stops at the first clear, valid zone rather than always running to the floor:
Pick the sector tab — Forex, Crypto, Stocks or Metals — to load its instrument list.
Type in the search box (e.g. “EURUSD” or “Euro”) and select from the dropdown. The list is filtered live as you type.
Only the timeframes allowed by your selected mode appear. The selected timeframe highlights in gold. (You must pick a mode first.) The mode you pick is passed to the engine with the timeframe, and each mode accepts only its own timeframes.
M3 and H8 are real three-minute and eight-hour candles, built from one-minute and four-hour bars because the market-data provider has no 3-minute or 8-hour interval. H8 candles open at 17:00, 01:00 and 09:00 New York time, the same bars most MT5 brokers show as 00:00, 08:00 and 16:00 server time.
Turn on “Macro Data Context” for a hybrid technical + fundamental read. When on, the engine also considers interest-rate & yield differentials, inflation and the broad macro backdrop, and it scans for scheduled high-impact news around your trade. When off, the analysis is purely technical.
Turn on “Sniper Entry” for a second, parallel precision read of the same setup: the engine relocates the entry into a micro point-of-interest on the refinement timeframe — measured on the leg a completed liquidity sweep produced, where there is one — with the stop and the targets inherited unchanged from the standard read. Only the entry is refined, so your risk falls because you got a better price — not because the invalidation was moved closer. The standard setup is always shown too. (See “Sniper Entry Refinement”.)
Press “Initialize Analysis Cycle”. It becomes available once a mode, instrument and timeframe are set. Running an analysis spends one Analysis unit. You will see a “Decoding Market Architecture…” state while it works.
Every analysis card has a “Copy post” button at the bottom. It copies the setup as a ready-to-post text: instrument, timeframe and mode; direction and order type; entry, stop loss and up to three take-profits; risk:reward, confluence and quality scores; and a two-sentence “why” taken from the analysis narrative — with emoji and bold headers that display on X. Sniper- or news-refined levels are used when the card shows them. On a no-trade result it copies the bias, the scores and the reason the setup was withheld instead of prices. The text always ends with the “Not financial advice” line, a link to athenixflow.app and the hashtags. Copying is free — it never spends a unit.
Every field in the analysis output, explained.
The result card (“Probabilistic Terminal Output”) packs the whole read into one view. Here is what each part means, top to bottom.
Total Confluence is scored out of 40 — the sum of four 0–10 sub-scores: Structure, Liquidity, POI and Premium/Discount. Below it sit three 0–100 reads: Quality (overall setup grade), Impulse (strength of the continuation setup) and Corrective (strength of the counter-trend setup). A total under 20 is flagged as below the reject threshold. (See “How Scoring Works” under Trade Setups.)
Three scenarios whose weights sum to 100: Scenario A — IRL reaction (price reacts at internal liquidity); Scenario B — IRL → ERL (price runs from internal to external liquidity); Scenario C — Expansion (price expands beyond the external range). The bars show how the engine weighted them against each other, and the heaviest one is emphasised.
The panel that shows how the setup was actually reached, rather than only where it ended up. It carries the selected point of interest — its type, the timeframe it was chosen on, its price zone, its score out of 50, whether it is unmitigated, whether it is the extreme valid zone, and the liquidity that justifies it — followed by the refinement ladder and the invalidation level.
Session Context gives the active session and killzone, the current New York time, the Asian session high/low, and whether either side of the Asian range has been swept. Market Story reconstructs the sequence: origin of the move → current phase → the projected liquidity path (which pools price is likely to take next).
These session facts are measured server-side from a real clock and your instrument’s candles, then handed to the engine — so the session named on the card, and in the written reasoning, is the session it actually is. Killzones are anchored to New York time and shift against GMT with US daylight saving.
Buy-side and sell-side liquidity pools are listed as price levels, along with a projected liquidity path and a liquidity heatmap describing where pressure/volatility is concentrated.
Volatility Context describes the current regime (e.g. compressed, expanding). The Reasoning grid explains the HTF structural framework and the liquidity engineering behind the read, in plain language.
Impulse vs corrective setups, entries/stops/targets, and what the scores mean.
Each setup shows a Buy/Sell direction badge and its price levels. If the engine finds no valid setup of a type, it shows a “No … Setup Detected” placeholder rather than inventing levels. The counter-trend setup is also put through its own stop floor and structural checks; when it fails them the card says “Counter-trend setup withheld” with the reasons, and shows no levels.
Confluence (out of 40) = Structure + Liquidity + POI + Premium/Discount, each scored 0–10. A total below 20 is a reject. Quality (out of 100) grades five dimensions (structure integrity, liquidity alignment, POI quality, session timing, risk-reward geometry); 70+ is the “valid” line. The verdict badge combines these: VALID needs a trade decision AND confluence ≥ 20 AND quality ≥ 70; otherwise you’ll see BELOW THRESHOLD or NO TRADE.
Where resting liquidity sits, scored, charted and explained.
Every analysis carries a Liquidity Map: a price chart with the pools drawn on it, above a price-ordered ladder that scores each one 0–100 and shows exactly why it scored what it did. It is measured from your instrument’s real candles across the execution timeframe plus daily and weekly, and it is available on every plan.
The shaded bands are liquidity pools — green above price, red below — and the stronger the pool, the more solid the band. Pools already swept are drawn as a dashed outline instead of a filled band. Your entry, stop and targets are marked as labelled lines, along with equilibrium and the true day open, so you can see the whole setup positioned rather than listed.
Pools above the spot marker are buy-side liquidity (buy stops resting above old highs); pools below are sell-side. Bar length and the number beside it are the significance score — how much the level matters if price gets there. Tap any row to see the factors behind it, and the measured chance of price reaching it.
A level that has already been taken is not the same as one that has not. Every pool carries a state, and a swept pool always ranks below an otherwise identical fresh one.
Crypto is the one market we support where a real order book exists and can be read. On crypto analyses we check the live book and mark any pool it corroborates with a “book” tag; a large wall with no structure behind it appears on its own as “book only”. Real resting size can lift a pool’s heat by up to 25%, and never lowers it.
Indices have no live price feed yet, so they show an explicit “no liquidity map” note rather than an empty grid or a blank chart.
The supporting, code-computed read that sits alongside the analysis.
Below the main output is the Structure Intelligence panel — a supporting read computed directly from price structure. It informs your view; it does not override the engine’s setup.
When each session runs, which pairs move in it, and what suits scalping vs day trading.
The market is not equally tradeable at all hours. Liquidity arrives in waves, and the same setup that works cleanly at the London open will drift sideways for three hours in the middle of the Asian session. The Market Session panel on your dashboard shows which session is live right now, the countdown to the next killzone, and the pairs that are actually moving.
Smart money accumulates in Asia, manipulates at the London open, and expands through New York. Most intraday setups are a variation on that sequence.
What the Macro toggle adds and how to use fundamentals with technicals.
With the Macro Data toggle on, the analysis gains a Macro / Fundamental Context panel drawn from real economic data (via FRED). The scheduled-news read — the release calendar, event phases and the News Risk level — is a separate layer and comes with every plan; see News Risk & the Release Calendar.
When a release the instrument answers to has printed inside the window, the panel adds a macro read. Each US print is fetched as it was first published (FRED/ALFRED, with the vintage and timestamp shown) and scored against the consensus the calendar carried: the raw surprise, the standardised surprise in units of that series’ historical surprise scale, a bucket from large miss to large beat, the prior as first printed and any revision to it, and — for a bundle such as the jobs report — its components. A print at the consensus, including a policy decision at the expected rate, is “in line”: not a surprise.
Best practice: use fundamentals for BIAS (which way and how strongly), and technicals for the ENTRY. When macro agrees with the technical structure, treat the setup as higher-conviction; when they conflict, lower conviction and favor the more conservative scenario.
The release calendar, the News Risk level, event phases and the news-refined setup.
Scheduled economic releases are the one thing a chart cannot show you in advance. AthenixFlow reads a calendar for eight currencies — USD, EUR, GBP, JPY, AUD, NZD, CAD and CHF — and attaches a scheduled-news read to every analysis, on every plan. It is a risk read, not a direction read.
Before you spend a unit, the News Risk panel under the Session panel shows the level for the instrument you have selected, live: LOW when nothing relevant is inside the trade horizon; MODERATE when a relevant release is further out, or a release for another currency is imminent (spillover); HIGH inside four hours of a tier-1 release the instrument answers to, or in its confirmation window; EXTREME when such a release is imminent, on the wire, in its first reaction, or when two of them land inside an hour. Each level names the release behind it.
At the start of each trading day the calendar is checked once, and if anything notable is scheduled a banner appears across the terminal. It names each release with its currency, its US-Eastern time and your local time, the consensus and previous where published, which session and killzone it lands in, and what to do about it — including the blackout window to avoid entering across (±15 minutes for high impact, ±5 for medium).
The banner only appears when there is something to say; a clear day shows nothing. Dismiss it and it stays hidden until the next trading day.
Each analysis carries the read it was taken under: the level and its reason, the mode (pre- or post-release), and each release inside the horizon with its currency, date, US-Eastern and local time, countdown, consensus, previous and phase. Tier-1 releases show in a warning tone; secondary releases as a lighter note.
With the Macro toggle on (Trader Pass or Pro Pass), a tier-1 release the instrument answers to also makes the engine produce a separate News-Refined Setup that anticipates the manipulation: the entry is placed at the zone price is likely to be driven into, the stop sits just beyond the expected sweep, and the take-profits are unchanged. It’s shown alongside the standard setup — the standard entry is never overwritten.
The precision-entry overlay: the entry is refined on the swept leg; the stop and targets are inherited unchanged.
With the Sniper Entry toggle on (Pro Pass), AthenixFlow refines the same trade idea into a precise, structure-anchored entry. The analysis engine authors it: it selects the point of interest, then relocates the fill to a smaller point of interest INSIDE that zone on the refinement timeframe — a refined order block, a micro fair value gap, or the origin of the displacement that followed a liquidity sweep.
An optimal-trade-entry is a retracement of a leg, so the leg has to be the right one. Where a liquidity sweep has completed and structure has shifted, the entry band is the 62%–79% retracement of the leg that sweep produced — from the wick that took the liquidity to the extreme of the move away from it — with 0.705 the precise level. That leg is measured server-side from real candles and handed to the engine, so it is not a guess.
A fibonacci level with no point of interest at it is never an entry on its own. If there is no order block or fair value gap sitting in the band, the refinement declines rather than filling at a bare price.
The liquidity map is a support layer, not a separate opinion. Before the analysis runs, AthenixFlow measures where liquidity rests — equal highs and lows, session and previous-period extremes, inducement, and which pools have already been swept — and supplies it to the engine. That evidence is what the engine scores its Liquidity Alignment and Heatmap Alignment against when it chooses the point of interest. It ranks the zone; the zone sets the entry. Liquidity never sets a price directly.
Inducement is nearer liquidity that exists to be taken first — the dip that fills large resting orders on the way to the real objective, and the move that collects stops parked on the obvious level. Where two candidate entries are otherwise comparable, the refinement prefers the one that is not sitting on an inducement pool. When every candidate is contested the setup is still shown, marked “Entry contested by inducement”, because resting liquidity beneath a discount entry is the normal state of a market rather than a fault — what protects the trade is where the stop goes.
The card shows each entry with the zone it came from, the blended average if both fill, the single stop and what it is anchored to, the unchanged targets, and projected risk-reward both for Entry 1 alone and for the blended fill, with the Entry 1 figure also shown net of the estimated spread. A “structure aligned” chip confirms the entry sits on the correct side of equilibrium — buys from discount, sells from premium.
The refinement fails closed rather than inventing a setup. It will decline when there is no valid point-of-interest to build from (a fibonacci level with nothing at it is not an entry), when the direction conflicts with premium/discount, when the refined entry is no better than the standard one, when it sits past the standard read’s own invalidation or at/beyond the first take-profit, when the inherited stop sits inside an imbalance or inside ordinary noise for the timeframe, when the refined entry sits outside the point of interest the standard read selected, when the standard setup itself was withheld, or when the read lacks displacement conviction. The reason is stated on the card.
Your analysis ledger, how forecast status is tracked, and recording results.
AthenixFlow grades each call by replaying real price, candle by candle, from the moment the read was created, on the timeframe you analysed. It checks exactly the entry, stop and three targets printed on the card. Every forecast sits in one state: Pending (entry not reached), Active (filled), TP1 or TP2 hit, Completed (final target reached), Stopped out, Cancelled (TP1 traded before the entry filled), or Expired, meaning the entry was not reached in time, the level it was built on broke, or the trade ran past its window: 12 hours for a scalp, 48 hours for a day trade, 10 days for a swing. Completed, Stopped out, Cancelled and Expired are permanent. Each touch is charged half an estimated spread. If one candle holds both a target and the stop, the engine checks a lower timeframe; if it cannot tell, it assumes the stop came first and labels that. A read that hit TP1 before stopping is recorded as hit TP1. NO TRADE reads are not tracked. Analysis and education only. Not financial advice.
Every analysis is saved to your analysis history. Each entry is a collapsible card (instrument, direction, date, key levels); expand it to see the full analysis again.
When two or more of your open forecasts share a currency and lean the same way on it, the Ledger says so above the list. Long EUR/USD and long GBP/USD look like two separate ideas, but both are really a bet against the dollar — if the dollar strengthens, both lose together.
Every forecast is always in exactly one state, worked out by replaying every candle since it was created. It is measured, not judged — the same forecast and the same price history always produce the same state.
Expand a saved forecast to see exactly what happened and when: entry fill, each target, and the stop, with the price and timestamp for each. The replay uses the timeframe you picked and grades exactly the entry, stop and targets the card shows. It starts from the moment the forecast was created, using finer candles for the first, partly-elapsed candle. If price touches a level on the candle that is still forming, the card says so in gold; that becomes final when the candle closes and you check again.
Tell the system how a trade actually went via “+ Log Trade Outcome.” Pick one outcome and (optionally) add a comment:
Logging is still useful for instruments the engine cannot price — those outcomes are recorded as self-reported and clearly marked. Where the engine CAN measure the price path, its result is authoritative and a logged outcome will not override it.
The outcomes you log also power the Trade Performance section on your Dashboard — a personal, at-a-glance read of how your AI-Assistant trades are working out. It summarizes only the trades you’ve logged, so it fills in as you record results.
Ask an SMC mentor anything, at your level.
The Education Hub is your personal Smart Money Concepts mentor. Ask any trading question and get a tailored, structured answer.
Every Q&A is saved to your Learning History (right column) as a collapsible entry showing the category, the exact learning depth used for that answer, and the date, so you can revisit answers anytime.
Log trades, track performance, and review outcomes.
The Trade Journal lets you record and review your trades — separate from analyses, this is your own performance log.
Filter your history by outcome, trade type and timeframe. Expand any trade to see its levels, date, notes and risk-reward; update the outcome as the trade develops, or delete an entry.
Setups posted by named publishers, with a live outcome status.
The Trade Setups feed shows setups posted by named publishers. It is optional — the AI Assistant works without it, and nothing here is financial advice.
The Starter Pack, the two passes, units and topping up.
All prices are in US dollars. On the Passes & units page you can choose to pay in a local currency where one is offered (for example naira): the page shows the converted amount, you are charged that amount at the day’s rate plus a small buffer for exchange-rate movement, and the checkout then offers the local payment methods available for that currency — such as bank transfer, USSD or mobile money — alongside cards. Your units, plan and receipts are always recorded at the US-dollar value, whichever currency you paid in. The first time you open the page you are asked where you are paying from; your profile country (or, failing that, your current location) is suggested. Pick your country and every price on the page — plans and token top-ups alike — is shown in that country’s currency where we accept it, with the US-dollar price underneath; elsewhere prices are in US dollars and the prompt says which local currency is coming soon. You can change the country at any time from the “Paying from” card above the plans, and switch to paying in US dollars there if you prefer. The card also lists the payment methods your currency unlocks on the payment page — for naira: bank transfer, USSD, NQR, Opay, eNaira and card; for Ghana, Uganda, Rwanda and other mobile-money countries: mobile money and card; for Kenya: M-Pesa and card. If a local currency cannot be priced at that moment the page says so, and paying in US dollars is always available. For some currencies we set fixed local prices instead of converting; the page then shows the exact amount you will pay rather than an estimate.
When Flutterwave sends you back to the Billing page, we check the payment against our own record before saying anything. A successful payment shows “Payment received” with your new plan or the units added; a declined or cancelled payment says so plainly — nothing has been charged, and you can try again or pay in another currency. Bank transfers can take a minute to confirm: if the page is still waiting after about thirty seconds it tells you the plan and units will appear once the confirmation arrives. Your Billing page always reflects the confirmed state; refreshing it never charges you.
For launch, the first 500 members to buy a pass with code ATHENIXLAUNCH get 50% off their first Trader Pass or Pro Pass. Enter the code on the Passes & units page before choosing a pass — the discounted price is confirmed at checkout. It’s one-time per account and stops automatically once the 500 spots are claimed, after which standard pricing applies. The code does not apply to the Starter Pack or to top-ups. The pricing and Passes & units pages show how many of the 500 places are claimed, and an account that used the code carries a “Founding member” label.
On the Passes & units page, buy the Starter Pack in one press, or choose Analysis or Education, enter an amount (minimum $5.00) and authorize the top-up. Rates: $5.00 = 20 Analysis units, and every Analysis top-up brings one Education unit per read with it (so $5 = 20 reads + 20 mentor questions); $5.00 = 100 Education units on their own. Units never expire, and units once used are non-refundable — each read is a real model call.
Give 2 reads, get 3 when they run one and 10 when they buy. Partners earn cash instead.
Every account has an invite link under “Invite a trader” in the sidebar (athenixflow.app/?ref=YOURCODE — the code is assigned by the server the first time you open the page and never changes). Anyone who signs up through it is attached to you, once, at signup; a code can never attach to its own account.
If you run a room or an audience you can apply to be a partner from the same page. Partners earn cash instead of reads: 20% of a referred user’s first purchase and 10% of every later purchase for 12 months — packs and passes alike — tracked in USD and paid out by bank transfer once a month. Approval needs a named room, no broker links, and the one-page agreement; a partner is disclosed as a paid partner on the landing page their link opens. The people a partner invites still get their 2 extra reads.
Your profile, system checks, and session control.
Settings shows your profile (name, email, access tier & status, join date, residence, and current Analysis/Education balances) and gives you system tools.
Pick your display language in Settings or from the globe in the header. The change is instant, is saved to your account so it follows you across devices, and covers the terminal interface and the public site. The AI mentor also answers in your language, and “Explain this analysis” explains a result in it.
From Settings you can leave a star rating (1–5) with an optional headline and a short review. Reviews are moderated — once our team approves yours, it appears publicly on the AthenixFlow homepage.
Quick answers and best practices.