Verify the installed module
Follow the install steps above, then run the export check from the dashboard root. Expected result: strategy export present. This checks loading only; it does not submit an order.
Loading...
Examine bands, RSI and trend context before treating an extreme as a reversal.
This is downloadable source software. Hosting, trading funds and external services are separate. Build and signal checks are not certification for unattended real-money execution. Complete venue and recovery validation for your installation. Some packaged agent INSTALL/manifest files retain the older word “candidate”. This guide is tied to the registered September 28 archive by checksum; publication does not certify venue performance.
Markets can stay at an extreme. Do not remove trend checks merely to increase trading activity. Middle-band targets can move as new data arrives; runtime protection and replacement still require independent verification.
Windows: Get-FileHash .\your-download.zip -Algorithm SHA256. Linux: sha256sum your-download.zip. macOS: shasum -a 256 your-download.zip.
node -e "const p=require('./plugins/mean-reversion-agent/dist/strategy.js'); if(!p.strategy) throw Error('strategy export missing'); console.log('strategy export present')"The module combines Bollinger Bands, Keltner Channels and RSI, with checks for narrow ranges, band walking and double-bottom/top patterns. It proposes a single target at the middle or opposite band; it does not submit staged one-third exits.
These defaults are a starting point for comparison, not recommended trading parameters or a loss limit. Candle counts depend on the interval. Refer to the shipped source for validation rules before changing them.
| Setting | Default | Meaning |
|---|---|---|
| bbPeriod | 20 | Bollinger Band SMA period |
| bbStdDev | 2 | Standard deviation multiplier |
| kcPeriod | 20 | Keltner Channel EMA period |
| kcMultiplier | 1.5 | ATR multiplier for KC |
| rsiPeriod | 14 | RSI period |
| rsiOversold | 30 | RSI oversold threshold |
| rsiOverbought | 70 | RSI overbought threshold |
| bandwidthThreshold | 5 | Squeeze threshold % |
| walkingThreshold | 3 | Periods at band to consider trending |
| doubleBottomLookback | 20 | Periods to check for double bottom |
| minBodyPercent | 0.3 | Min candle body % for valid signal |
| takeProfitTarget | "middle" | TP at middle band or opposite band |
Work in an isolated paper or research environment. The exercise teaches installation and signal interpretation; it does not send an order.
Follow the install steps above, then run the export check from the dashboard root. Expected result: strategy export present. This checks loading only; it does not submit an order.
Select 15m, 1h, 4h as supported by the manifest. Supply one completed candle series for one symbol, retaining timestamps, OHLCV, source and network. Multiple listed intervals do not mean automatic multi-timeframe fetching.
Use completed 1h candles, bbPeriod=20, bbStdDev=2 and takeProfitTarget="middle". Compare a ranging sample with a persistent move along a band. Record the RSI, band position and reason for the returned decision.
A touch of the lower band is not automatically a buy. Read the regime and confirmation checks. bandwidthThreshold=5 is expressed in percentage points in the source calculation. Compare target choice only in a separate run.
Confirm ownership, sizing, fees, entry fill, stop/target, modifications, exit and realized P&L on testnet. Include restart, partial fill, rejected cancellation and lost-response reconciliation. A signal test is not an execution test.
Narrated, captioned lessons showing the broader demo dashboard. Some screens and hosted workflows differ from the purchased source edition. Use this product’s written installation and practice tutorial for the exact release.
A touch of the lower band is not automatically a buy. Read the regime and confirmation checks. bandwidthThreshold=5 is expressed in percentage points in the source calculation. Compare target choice only in a separate run.
Check order status in Account and contact support with the order reference if fulfillment is delayed. Do not make another purchase to repair access.
Generate a fresh link from the purchasing account. The documentation downloads are public; the paid software archive remains entitlement-protected.
Confirm manifest.json is directly inside the manifest installPath, the entryPoint exists, compatibility matches and the loader has no errors. Restart after installation; do not add another nesting level.
Inspect the latest decision, candle history and freshness, symbol, interval, network, existing positions and risk rejection. Neutral is a valid result.
Keep the exact failed command and redacted diagnostic. Check Node, psql, Supabase settings and owner verification. Do not bypass types, authentication or schema integrity checks.
Compare the same account and network at the venue, including fills, fees, margin and open orders. Reconcile before retrying; a lost response does not mean the order failed.
Include your order reference, product/version, operating system, Node version, failed step and a redacted error. Never send environment files, passwords, private keys or seed phrases.