Kryll Guide

This lesson covers one of the 5 platforms that I can recommend for algorithmic trading. It requires no coding, easy to use, but only supports crypto.

Let's get started.


Part 1. Quick-Start

Check this official guide with videosarrow-up-right

  1. First, you need to have an exchange ready. It can be Coinbase, Binancearrow-up-right, KuCoin, Kraken, Gate.io, Crypto.com, Bittrex, or HitBTC.

  2. Create a free Kryll.ioarrow-up-right account. Username, email, password — as always.

  3. Fill the investor profile. Pick Long‑term and Security for low‑risk defaults.

  4. Link the exchange by API key.

    • Spot‑trading and “Read/Trade” only.

    • Paste keys into Kryll.

    • Receive $7 KRL credit.

  5. Enable 2‑factor auth and set a security question.

  6. Launch the first bot (Marketplace).

    1. Click a bot with solid stats.

    2. Hit Launch.

    3. Choose the linked exchange.

    4. Select one trading pair.

    5. Allocate at least $200.

    6. Press Start.

  7. Watch it work. Open My strategies → Active to track fills, PnL, and fees.

  8.  Install the Kryll app to monitor bots, swap coins, and launch new strategies on the go.


Part 2. First Steps in Editor

Time to move from “watching” to “building”.

First, get inside. Log in, then hit Strategy Editor on the left. Click Create a new strategy. You’re now on the workbench.

1 . Know the four zones

Zone
What it shows
Why you care

Settings bar

Pair, timeframe, back‑test window, balance, Pro‑mode toggle, auto‑save.

Swap pair fast, tweak back‑test dates, switch to TradingView chart if you want indicators.

Chart

Candles, indicators (Pro‑mode), mock fills when you back‑test.

Visual proof the bot is firing where it should.

Block list

All indicator blocks and logic blocks.

Your toolbox. Every strategy is just blocks wired together.

Editing area

Empty grid—the canvas.

Drag blocks here, set them up, link them into a flow.


2 . Handle blocks like a pro

Action
What to do

Drop a block

Click‑hold a block in the list, drag to the canvas, release.

Move

Click‑hold a block, drag, drop.

Select

Double‑click a block.

Open settings

Double‑click the thin bar under the block.

Duplicate

Select, press D. All params copy over.

Delete

Select, hit DEL or BACKSPACE.

Link

Select the first block (double‑click), then click the next block. Flow arrow points from first to second.

Pick a link

Double‑click the line, or re‑select the two blocks.


3 . Fast keys you’ll use every day

Key
Result

Z

Undo last action.

Y

Redo.

SPACE + drag

Pan the whole canvas.

SHIFT + drag new block

Auto‑link that block to the last one you placed.


You can now drag, drop, wire, and tweak blocks without thinking. In the next part we’ll build a real strategy and run the first test.


Part 3. Build your first bot (SuperTrend)

We’ve clicked around long enough. Let’s wire a live strategy.

1 . Prep the canvas

  1. Turn on Pro‑mode – click Pro on the settings bar to swap in the TradingView chart.

  2. Add SuperTrend – top‑left of the chart press Indicators → SuperTrend (Kryll).

  3. Switch chart to 1 D; that’s the timeframe we’ll trade.

2 . Drop the logic

  1. Drag SuperTrend from the block list to the canvas. Open its settings, set timeframe to 1 D, keep direction = Bullish, close.*

  2. Notice the lone Wallet block already there—it’s your starting node.

  3. Link Wallet → SuperTrend (double‑click Wallet, click SuperTrend). Flow arrow appears.

  4. Drag in a Buy Limit block. Link SuperTrend → Buy Limit. Now the bot buys when SuperTrend flips green.

3 . Add the exit

  1. Select the SuperTrend block, hit D to duplicate.

  2. Open the copy, switch direction to Bearish.

  3. Drag in a Sell Limit block. Wire Bearish SuperTrend → Sell Limit.

So far: Wallet → Bull ST → Buy → Bear ST → Sell.

4 . Make it loop

Right now it’s one‑shot: after that Sell it dies. Two quick ways to loop:

Option
How

Link back

Simply connect Sell Limit → Wallet. Straightforward, can get messy later.

GoTo block (cleaner)

Drag GoTo, set destination = Wallet, close. Link Sell Limit → GoTo. Press G any time to reveal hidden GoTo links.

Pick one; same result.


5 . Backtest

  1. Top bar: choose Binance – BNB/USDT.

  2. Set start 2018‑01‑01 and end 2020‑12‑31.

  3. Hit Test.

A progress bar crawls; bigger date ranges take longer. When it finishes the Analysis pane shows:

  • Net % PnL

  • Trades count

  • Max drawdown

  • Equity curve on the chart

You should see multiple round‑trips now that the loop is in place.

6 . What next?

Scroll the chart. Each green triangle = buy, red = sell. If fills don’t line up with SuperTrend flips, revisit block settings.

Save the strategy; we’ll upgrade it with stops and risk rules in the next part.

You just built, looped, and tested a working bot without writing a single line of code. But we are not there yet.


Part 4. Versioning, Workbench, and Deep Analysis

By now you’ve got a looping SuperTrend + Stoch RSI bot. Let’s stress‑test it, track every tweak, and find out if it’s worth real money.


1 . Upgrade the logic

  • Pair & TF – keep BNB/USDT, 1 D.

  • Blocks

    • Drag an AND block onto the canvas.

    • Into slot 1 drop SuperTrend (Trend = Bullish, TF = 1 D).

    • Into slot 2 drop Stoch RSI (Signal = Overbuy, TF = 1 D, Overbuy level = 50).

    • Link AND → Buy.

    • Duplicate Stoch RSI, set Position = Neutral. Link to a Sell‑50 % block.

    • Duplicate SuperTrend, set Trend = Bearish. Link to a final Sell‑100 % block.

    • Close the loop with GoTo → Wallet.

This stack buys when trend flips up and Stoch RSI sits above 50, scales out when momentum cools, and exits on trend reversal.

2 . Create a test suite

My rule: diversify the back‑test or the bot will ambush you live.

  1. Hit Test with BTC/USDT first — start 2021‑04‑14, end 2023‑01‑24.

  2. When the run finishes, press Workbench → Add. That snapshots code and stats.

  3. Repeat for ETH, BNB, SOL, FTM, DOT, MATIC, LUNA (for the laugh), CAKE, THETA, ENJ. Ten pairs, same dates.

Now the Workbench holds ten sealed results you can open with one click.


3 . Use versioning like Git for traders

  • After every meaningful change — new exit rule, different Stoch level—hit Save → New version.

  • Kryll stores the code diff plus the test history.

  • Need to roll back? Just select an older version in the dropdown and you’re instantly there.

  • I tag versions “v2‑half‑exit” or “v3‑tight‑stop” so I know what changed without opening the flow.


4 . Read the analysis window

  • Perf % – net growth on that pair.

  • Trades – more isn’t better; sweet spot is 5‑30 per year on daily charts.

  • Equity DD vs Max DD – equity is intra‑trade pain, max is worst closed loss. Keep equity DD under 25 %.

  • Charts – hover the equity curve; look for long flat spots or V‑drops.

When the surface stats look solid, smash Deep Analysis at lower right. It drills into:

  • Per‑trade heat map.

  • Monthly breakdown (find dead months).

  • Risk metrics: Sharpe, Sortino, Win/Loss size.


5 . My personal filters

I green‑light a bot only if across the ten‑pair basket:

Metric
Threshold

Median equity drawdown

≤ 25 %

Max pair drawdown

≤ 50 %

Sharpe

≥ 1.0

Annualised return

≥ 40 %

Losing streak

< 8 trades

Our test set hit 152 % total, ~67 % CAGR, equity DD ‑22 %, max DD ‑46 %. Good start, but Sharpe was soft. I’d tighten stops or add a time‑exit before letting it touch real capital.


6 . Tips from 80+ live bots

  1. Save first, experiment later. Always clone a new version before a drastic tweak.

  2. Workbench isn’t a trophy case— clean it. Keep only the top five snapshots; clutter hides flaws.

  3. Test baskets, not heroes. A bot that crushes one coin and bleeds nine others is a landmine.

  4. Walk‑forward every quarter. Shift the test window 90 days and rerun; if performance collapses, the edge is gone.

  5. Document in the version notes. Future‑you won’t remember why “v7‑rev‑B” exists.

Lock these habits now and you’ll iterate faster than most desk‑bound coders.

Now you can create your trading bots on Kryll. This guide was relatibly simple (we only launched a basic strategy bot). Next, you need to spend 5-10 hours to try different tools and approaches to master this tool and start creating something more complex. You may do it now, before you move on to the next tool. Take your time.

Last updated