Quickstart

🦉 AthenaAI is an AI-powered ensemble machine-learning trading bot for PocketOption, built on BinaryOptionsToolsV2 → Get it on GitLab
💬 Need help? Join the Chipa Discord · ⚡ Go no-code with ChipaEditor

This guide takes you from an installed AthenaAI to a running bot in five steps.

1. Get your PocketOption session ID (SSID)

The bot authenticates with your PocketOption session ID, read from the PO_SSID environment variable. To obtain it, log in to PocketOption in your browser and extract the session ID from the site’s websocket authentication (browser DevTools → Network → WS frames; the session string is what BinaryOptionsToolsV2 expects as ssid).

Never put the SSID in source code — the bot only reads it from the environment or .env. If it is missing, main.py exits with:

ERROR: PO_SSID is not set.
Provide your PocketOption session ID via the environment:
  export PO_SSID='your-session-id'   # Linux/macOS
  $env:PO_SSID='your-session-id'     # PowerShell
or copy .env.example to .env and fill it in.

2. Create your .env

Copy the template and fill in your values:

cp .env.example .env      # Linux/macOS

Copy-Item .env.example .env   # Windows PowerShell

The template covers the required SSID plus the most common trading, money-management, and persistence settings:

PO_SSID=your_session_id_here
PO_ASSET=EURUSD
PO_TIMEFRAME=60
PO_EXPIRY_OPTIONS=120,300,600
PO_DEFAULT_EXPIRY=300
PO_BASE_STAKE=25
PO_MAX_STAKE=100
PO_MAX_DAILY_LOSS=300
PO_MIN_CONF=0.63
PO_MAX_CONF=0.85
PO_REQUIRE_ALIGNMENT=1
PO_DATASET=EURUSD_M1.csv
PO_BRAIN_PATH=athena_brain.pkl

main.py loads .env with os.environ.setdefault, so a variable already set in your real environment wins over the .env value. Details in Configuration; each variable has its own page under the environment reference.

3. (Optional but recommended) Train a brain first

Train offline from historical data — no PocketOption connection needed:

python train.py

Defaults train on EURUSD_M1.csv and save to athena_brain.pkl, which matches the .env above, so the bot loads it at startup and skips in-process pre-training. A successful run ends with a walk-forward report and:

✅ Training complete — brain saved to athena_brain.pkl

See the full Training guide. If you skip this step, the bot pre-trains itself from PO_DATASET at startup instead (slower, done in-process).

4. Run the bot

python main.py

What you should see

The startup banner (real log lines from src/bot.py):

════════════════════════════════════════════════════════════
  🦉 AthenaAI TRADING BOT v6.0 — PocketOption
  Asset: EURUSD  |  Timeframe: 60s
  Expiry: AI-selected from ['120s', '300s', '600s']
  Models: SGD + PA + NB + GBM(0.75×) + RF(0.75×)
  Confidence: 63%–85% | Hours: 0,1,2,3,4,16,17,19,20,21,22,23 UTC
  Features: 40 core + 17 experimental  |  Adaptive: ON
════════════════════════════════════════════════════════════

Then one of two paths:

• Saved brain found:

  🧠 Brain loaded (fitted=True, batch=True, models: {...})
  ✅ Loaded saved brain — skipping dataset training!

• No brain: the bot reloads the trade journal (🔄 Reloaded N trades from journal — models retrained! if you have history) and pre-trains from the dataset (Loading dataset from EURUSD_M1.csv …, 📚 Training on N samples (70%) …, a walk-forward report, and ✅ Pre-trained on N samples!).

Then the connection and warm-up:

Connecting to PocketOption …
Connected!  Balance: $123.45
Loading 60 warmup candles …
Loaded 60 candles.  Starting main loop …
📡 Using subscribe_symbol_time_aligned (timeframe=60s)

From here the bot polls for a trade opportunity every 2 seconds. While gates are blocking, it logs a ⏸ diagnostic line at most every 30 seconds (e.g. ⏸ Low confidence: 58.2% (need 63.0%) dir=call regime=ranging). When a trade fires you see a ▶ TRADE line, and later a ✅ WIN / ❌ LOSS result line. Reading these is covered in Live trading.

5. Stopping the bot

Press Ctrl+C. main.py catches the interrupt and runs a clean shutdown:

Interrupted, shutting down...
🧠 Brain saved to athena_brain.pkl
Bot stopped.  Final stats: ...

stop() saves the brain (and the companion athena_brain_expiry.pkl expiry stats) before exiting, so learned state survives restarts. Trade history is already persisted continuously in trade_journal.db — see Persistence.

Risk note

This bot places real trades with real money once connected. Binary options trading carries a high risk of losing money, and past performance (including walk-forward test results) does not guarantee future results. Start with a demo account and the smallest stakes your account allows, and set PO_MAX_DAILY_LOSS conservatively.

Next steps

• Configuration — all three configuration layers.
• Live trading — day-to-day operation and reading logs.
• Troubleshooting — if something above didn’t match.

💬 Stuck during setup? The Chipa Discord is the fastest place to get help — and if you’d rather skip the code entirely, build your strategy visually with ChipaEditor.