↻ loop← Back to the bots
THE LOOP PLAYBOOK

Your first scan.
One step at a time.

Supported: Windows, macOS and Linux with Python 3.10+. These instructions apply to Momentum Paper v1.1. It simulates orders and never connects to a wallet.

01 · Install on your computer

  1. Get the bot ZIP from your purchase delivery. Checkout is currently awaiting connection. Double-click it in Finder. Open the extracted pump-paper-bot folder. You should see bot.py and dashboard.html.
  2. Open Terminal from Applications → Utilities, or use Spotlight.
  3. Type cd with a space. Drag the extracted folder into Terminal, then press Return. This makes Terminal use the correct folder.
  4. Check Python:
    python3 --version
    Version 3.10 or newer is required. If unavailable, install Python from python.org, reopen Terminal, and repeat the folder step.
  5. Start the synthetic demo:
    python3 bot.py --demo
  6. Open http://127.0.0.1:8765 on that same computer. The heading should show v1.1 and SYNTHETIC DEMO. The artificial token demonstrates a buy and sell; its gains are fabricated test inputs.
Windows · PowerShell
  1. Extract the ZIP using Explorer → Extract All. Open the extracted folder containing bot.py.
  2. Install Python 3.10+ from python.org if needed. Include the Python launcher during installation.
  3. Right-click inside that folder and choose Open in Terminal. Check py -3 --version.
  4. Run py -3 bot.py --demo. Open http://127.0.0.1:8765 in Edge, Chrome or Firefox.
  5. Press Control+C, then run py -3 bot.py for real-data paper mode. If the launcher is unavailable but Python is installed, try python bot.py.

Alternatively double-click Start-Windows.bat for market-data paper mode. A console remains open; close it to stop.

Linux · Terminal
  1. Extract the ZIP. Install Python 3.10+ using your distribution's package manager if it is not present.
  2. Open Terminal in the folder containing bot.py.
  3. Run python3 --version, then python3 bot.py --demo.
  4. Open http://127.0.0.1:8765 in your browser.
  5. Press Control+C, then run python3 bot.py for market-data paper mode.

02 · Use actual market data

  1. In Terminal, press Control+C to stop the demo.
  2. Start market-data paper mode (on Windows, use py -3 bot.py):
    python3 bot.py
  3. Refresh the dashboard. Confirm MARKET DATA / PAPER ONLY.
  4. Keep Terminal open and your computer awake. Check that Last successful scan updates and that Discovery decisions populates.

No trades can be normal. There is no expected waiting time or guaranteed trade frequency. The bot only simulates entry when every rule passes.

You can use Start-Windows.bat on Windows or run sh Start-Unix.sh on macOS/Linux for market-data paper mode. Demo and market-data accounts are separate. The current program has no flag that enables real-money trading.

03 · Configure your rules

Create a plain-text config.json file in the bot folder. For example:

{
  "position_usd": 3,
  "max_positions": 2,
  "daily_loss_usd": 6
}

Stop and restart the bot after changes. Use a plain-text editor, not a Word document. Keep the exact .json extension.

DefaultMeaning
$100 paper cashInitial account balance; does not reset existing state
$5 per entry; 3 positionsMaximum initial allocation $15 at once
$20,000 liquidityMissing liquidity is rejected, not treated as verified zero
$3,000 5m volumeMinimum recent reported transaction volume
15 buys; ratio 1.5×Transaction counts, not unique wallet counts
+2% to +20% changeAllowed 5m reported price movement
15–1,440 minutesAllowed pool age, not token creation age
−10% / +20% / 30 minutesExit triggers based on estimated net proceeds
$10 daily lossEstimated equity trigger, resets midnight UTC; not guaranteed loss protection
1% fee + 1% slippage per sideAssumptions; incomplete model of live costs
30-second polls; 60-minute cooldownEntries are selective, not rapid-fire sniping

Add a watchlist

Open watchlist.txt and put one full Solana token mint address per line. Do not use a ticker or pool address. The bot rereads it each cycle. Only supported Pump.fun/PumpSwap venue pairs will be considered.

Keep or restart a paper account

paper-state.json stores your account and trades. Keep it when updating. For a separate account without deleting old data:

python3 bot.py --state new-paper-state.json

Pause vs stop

Pause new entries blocks buys but leaves simulated exit monitoring running. Control+C stops the whole process, including exit monitoring. Resuming cannot undo missed price movements.

04 · Troubleshooting

No tokens or trades appear

Read Last successful scan and the error text first. If scans are current and candidates appear, review their reasons. Missing liquidity and strict entry filters often block purchases. Do not infer that the strategy works because it makes more trades.

“Can't open file bot.py”

Terminal is in the wrong folder. Type cd , drag the folder containing bot.py into Terminal, press Return, then run the command again.

The dashboard will not open

Confirm the bot is still running in Terminal. Use http://127.0.0.1:8765, not HTTPS. The link only works on the computer running the program.

“Address already in use”

Another process is using that port. Stop the earlier bot Terminal with Control+C, or run python3 bot.py --port 8766 and open http://127.0.0.1:8766.

API error or discovery warning

Check internet access. A source warning means that discovery source failed. A quote-batch failure prevents the scan from placing orders. Old open positions are retained. Stale marks are estimates, not valid live exit prices.

My dashboard still looks like the old version

Stop the bot. Replace bot.py and dashboard.html in the folder Terminal actually uses, preserving account files. Restart and refresh the browser. The title should show v1.1.

Does sleeping my Mac stop it?

Yes, sleep can suspend scanning. You can run caffeinate -i python3 bot.py to inhibit idle sleep while it runs. Closing the lid can still interrupt execution.

05 · What this version does not do

No X sentiment, holder concentration, unique-buyer analysis, mint authority audit, wallet clustering, trade signing, real orders or 24/7 cloud service is included. Discovery samples are incomplete and paid boosts are promotional. Slippage, liquidity, execution delay, failed transactions, transfer restrictions and network costs are not realistically modeled.

Stop-loss orders are simulated triggers, not protection from a rug pull. Paper results cannot be represented as live returns. The default thresholds are unvalidated strategy settings.

Never paste a wallet seed phrase or private key into this site or send it to anyone claiming to provide support.

Data adapter documentation: DEX Screener →