Everything you need to run CORVUS.

From zero to a behavioral profile in under five minutes.

CORVUS only needs one thing to work: your trade history. The faster you get it in, the faster you see your patterns. Three paths, depending on what you have today:

1. 01Drop a CSV from your broker (any of 40+ supported formats — we auto-detect).
2. 02Paste from a spreadsheet — the importer maps columns interactively.
3. 03Add trades manually one by one (with screenshots) if you prefer the journal-first workflow.

The Free plan gives you 200 deals and 50 media uploads — enough to run Trade Analysis and Risk Assessment end-to-end. The four heavier behavioral modules unlock on Pro.

Sign up with email + password. We don’t require a phone number, social login, or KYC. Your account is provisioned in under a second and starts on the Free plan automatically.

1. 01Open /login and switch to the REGISTER tab.
2. 02Enter your email and a strong password (min 12 chars, mixed case, one symbol).
3. 03Confirm via the magic link we send you. The session is valid for 30 days on this device.

The importer lives at Dashboard → Deals → Import CSV. Drop a file (up to 2 MiB / 1,000 rows), confirm the column mapping we auto-detected (or pick a different broker profile), then commit. Each row is validated against the same rules as the manual form, and failed rows come back with a specific reason — fix in your CSV and re-import.

Broker profiles

Profile detection runs on the header row alone, so it works for any CSV that resembles a known broker’s export. If your broker isn’t in the registry, leave the dropdown on — Custom mapping — and the wizard’s synonym matcher will guess (open_time → opened_at, symbol → asset, etc.). Today’s registry:

• MetaTrader 5 — Positions CSV: Account History → right-click → Save As → CSV. Round-trip rows with both open and close columns. Distinguished from MT4 by the Position and Volume columns.
• MetaTrader 4 — Statement CSV: Account History → Save as Report → CSV (or a third-party CSV converter from the HTML report). Distinguished by the Ticket and Size columns.
• cTrader — Statement CSV: History → Export → CSV. Round-trip rows with explicit Entry Time + Closing Time columns.
• NinjaTrader — Trade Performance CSV: Strategy Analyzer / Trade Performance → export. One row per round-trip.
• TradingView — Strategy Tester: List of Trades → Export. Note: each round-trip is split across two rows (Entry + Exit sharing the Trade #); the importer treats each row as one deal — expect 2× row count. Row-pair aggregation is on the roadmap.
• CORVUS Generic Journal: the format the exporter produces. Matches if your CSV uses our canonical column names (opened_at, asset, direction, result, r_factor, …).

CSV format

We accept any reasonable CSV. If your broker isn’t in the auto-detect list, the manual mapper covers it. Headers are case-insensitive and fuzzy-matched.

time,symbol,side,qty,price,fees,exit_time,exit_price
2026-06-20T09:31:14Z,ESM6,LONG,2,5421.25,2.40,2026-06-20T10:04:02Z,5428.75
2026-06-20T11:02:47Z,NQM6,SHORT,1,19880.50,1.20,2026-06-20T11:18:30Z,19852.25

Exchange-native imports

Binance, Bybit and OKX exports drop in as-is. For MT4/MT5, export the account history HTML from the Terminal panel — we parse it directly.

As soon as the import completes, the worker queues a behavioral scan. You’ll see modules light up one by one in the Overview HUD — usually all done within 5–15 seconds for 1,000 trades.

1. 01Watch the SCAN ITER counter tick on the Overview header.
2. 02Each module surfaces its top finding as soon as it's ready.
3. 03Drill into any finding to see the trades behind it.

Free · Always on.

The Trade Analysis module ingests, normalizes and indexes every executed trade in your workspace. It’s the foundation every other module reads from. You’ll mostly interact with it through the Deals view, the Journal calendar and the Overview HUD.

What it does

• · Auto-pairs entries and exits into full round-trips
• · Attributes P&L to spread, slippage and commission
• · Builds the searchable deal index used by every other module
• · Renders the deal-level replay view

Pro · Coming Q3 2026.

Unsupervised clustering over your trade metadata — session, instrument, hold time, outcome, emotional context. Surfaces the clusters that have the strongest edge and flags the ones that decay.

Pro · Partial — revenge trading live; rule engine coming.

Define your rules in Settings → Trading Rules. Every new trade is checked against them and any violation is flagged with an estimated P&L impact. Weekly digest emails roll up the damage and show whether the trend is improving.

Example rules

• · Max risk per trade: 1% of account
• · No trading between 22:00–04:00 UTC
• · Max 5 trades per day
• · No re-entry within 5 min of a stop-out

Pro.

Infers emotional state from execution metadata: re-entry latency after losses, abnormal trade clustering, position size deviations from your baseline. Optional self-reported mood input from the journal sharpens the signal.

Pro · Edge breakdown live; regime detection coming.

Edge breakdown across asset, direction, session, hour-of-day and R-bucket is live today. Per-tag setup ranking and drawdown attribution per failure mode are live. The market-regime tag (trending / chop / high-vol) needs external market-data integration and is on the roadmap.

Free · Always on.

Composite live risk score (0–100). Updated on every new trade. Combines concentration, leverage, rolling drawdown and recent rule violations into one number. Set thresholds in Settings to receive a notification when the score crosses a level.

The Journal is your reflective surface — a calendar heatmap colored by outcome, per-day notes, screenshots and tags. Click any day to drop into the day view with all trades and the running P&L overlay.

Tags and setups

Tag trades with your own setup names. The Pattern Recognition module uses tags as labels when you provide them — otherwise it generates its own cluster names.

The Analytics tab gives you the statistical surface: equity curve, drawdown waterfall, return distribution, expectancy by setup, Monte Carlo simulation (10,000 paths by default) and regime detection.

Monte Carlo

Resamples your trade-level returns with replacement to project the distribution of equity outcomes over a forward window. Useful for stress-testing the “bad path” for sizing decisions.

The Settings page covers profile, password, notification thresholds, trading rules, journal preferences (mood scale, default tags) and data export.

Data export

Pro accounts can export every dataset (deals, modules, analytics snapshots) as CSV or JSON from Settings → Data → Export. Exports are generated server-side and delivered as a signed download URL valid for 24 hours.

We bill in crypto. Pick a duration tier (1 week / 1 month / 3 months / 1 year / lifetime), confirm the chain, and we generate a one-time deposit address. Plans activate on first on-chain confirmation.

Invoice lifecycle

• · Address generated and locked for 30 minutes
• · Deposit detected by the chain monitor
• · Plan activated automatically — usually under 60 seconds
• · Invoice + receipt saved to your billing history

Planned chains: BTC (mainnet), ETH (ERC20), USDT-TRC20, USDT-BEP20, SOL (mainnet). Each invoice generates a fresh, single-use address — addresses are never reused across invoices.

Full refund within 7 days of your first payment, no questions asked. After 7 days we offer prorated refunds for unused full months on monthly / quarterly / annual plans. Lifetime refundable within 14 days.

Pro · REST · JSON · Live.

The CORVUS REST API gives Pro users programmatic access to deals, analytics snapshots and behavioral findings. Generate a personal access token in Settings → Developer.

Authentication

Pass your token as a Bearer header on every request. Tokens are scoped to a single workspace and rotatable from Settings.

curl https://corvusanalytics.org/api/v1/deals \
  -H "Authorization: Bearer corvus_pat_••••••••••" \
  -H "Accept: application/json"

Core endpoints

• GET /v1/deals — list deals with filters
• POST /v1/deals — create a deal
• GET /v1/analytics/equity — equity curve series
• GET /v1/modules/{id} — latest module output
• POST /v1/imports — upload a CSV import

We POST a JSON payload to your https:// endpoint when a subscribed event fires. Every delivery carries X-Corvus-Signature: sha256=<hex> — HMAC-SHA256 over the raw body bytes, signed with the secret you received at create time. Compare with hmac.compare_digest (or your language equivalent) before processing.

Available events

• deal.created — a new deal lands in your journal (manual entry or CSV import).
• subscription.activated — a paid plan activates (initial purchase or renewal).
• webhook.test — synthetic, fired only by the “Test” button in Settings; useful for receiver connectivity checks.

Delivery headers

• X-Corvus-Event — event type (e.g. deal.created).
• X-Corvus-Event-Id — stable per-event UUID; reused across retries so receivers can dedupe.
• X-Corvus-Delivery — per-attempt id (different on each retry).
• X-Corvus-Attempt — integer 1..5.
• X-Corvus-Signature — sha256=<hex>.

Retries

Any non-2xx response (or network failure) triggers a retry. Five attempts total at 30s, 5min, 30min, 4h, 24h (exponential). After the fifth failure the delivery is marked failed_final — the endpoint stays active, you’ll see the failure in the deliveries panel.

# Receiver-side signature verification (Python)
import hmac, hashlib

def verify(secret: str, body: bytes, header: str) -> bool:
    expected = "sha256=" + hmac.new(
        secret.encode(), body, hashlib.sha256
    ).hexdigest()
    return hmac.compare_digest(expected, header)

120 requests / minute per token on read endpoints, 30 / minute on write endpoints. Limits are returned in standard X-RateLimit-* headers. Burst capacity of 2× the per-minute limit applies to short spikes.

By using CORVUS you agree that the service is provided “as is”, that CORVUS does not give investment advice, and that all trading decisions remain your own. Behavioral findings are statistical observations, not recommendations.

Full terms are available on request and will be linked here in plain PDF before launch. Until then, write to [email protected].

We collect: account email, password hash, your uploaded trade data, billing addresses for invoices you create. We do not collect: phone numbers, KYC documents, browsing history, third-party tracking beacons.

Your trade data is encrypted at rest and in transit. We never sell, rent or share it with anyone. Period.

All data processing is performed inside the EU (Frankfurt region). We act as a data processor for the trade data you upload, and as a data controller for the minimal account-management data we store.

A signed DPA is available on request for business accounts. Contact [email protected].

• v2.0 · 2026-06 · CLOSED BETA
  Operator console redesign. Three behavioral modules live (Emotional, Strategy Weakness, Risk + Monte Carlo). Mistake Detection partial (revenge trading only). Crypto billing on testnet (BTC + USDT-TRC20). Public marketing pages shipped with honest live/partial/coming statuses.
• v1.4 · 2026-05
  Strategy Weakness module out of beta. Monte Carlo simulation added to Analytics. Mobile-friendly journal calendar.
• v1.3 · 2026-04
  Emotional Analysis module GA. CSV importer for 18 new broker formats. Per-trade screenshot upload to MinIO.
• v1.2 · 2026-03
  Pattern Recognition module GA. Mistake Detection rule engine. API access for Pro accounts.