playbook

Bitkub Trade Playbook

Long-only swing trading on Bitkub THB pairs via the gokub-mcp tools. The tools are the hands; this is the brain. Follow the flow, respect the guardrails, never skip the validation gate.

Execution discipline (never violate)

• Run the full flow autonomously. Never pause mid-flow to ask whether to continue. Screen → regime → signal → size → validate → decide — all in one pass.
• Report results, not options. End with a decision table (pass/fail per coin, with entry/stop/TP for any that pass) and the reason. Never end with "ถ้าต้องการจะรันต่อ…" or any conditional offer.
• No raw data dumps. Never return raw candle arrays or large result blobs to the user. Summarize each intermediate step in ≤ 1 line.

Hard guardrails (never violate)

• Risk ≤ 2% of equity per trade. Size from the stop, not from a fixed amount.
• Long only, uptrend only. No shorting, no counter-trend entries.
• No entry unless validate_trade_setup returns can_trade=true. This is the gate.
• Take-profit ≥ 2R. If R:R < 2, skip the trade.
• Bitkub has NO native stop/OCO. Protect positions with client_side_stop_worker (poll it on a loop until triggered=true). Never assume a resting stop exists.
• Never lower a trailing stop. Move it up only.
• No averaging down on a short-term plan.

Key mechanic: indicators need candles

The analysis tools (check_market_regime, calculate_atr, calculate_rsi, detect_breakout_signal, detect_pullback_signal, calculate_ema, calculate_roc) take arrays (prices or OHLCV candles), not a symbol. So always:

1. get_historical_candles(symbols=[symbol], format="close") — returns result.prices[] directly.
2. Pass result.prices into price-based tools (EMA, RSI, ROC, regime) or use result.candles (default format) for OHLCV tools (ATR, breakout, pullback, exit signals).

Use the big timeframe (1D / 240) for regime + relative strength, the entry timeframe (60 / 15) for ATR + entry signals.