🎁 Get the FREE AI Skills Starter Guide β€” Subscribe β†’
BytesAgainBytesAgain
πŸ¦€ ClawHub

Pine Script Strategy Builder

by @shailendartigga16-collab

Write, fix, test, and backtest Pine Script v5 strategies and indicators for TradingView. Use when asked to create trading strategies, indicators, Pine Script...

Versionv1.0.0
Downloads380
TERMINAL
clawhub install pine-script-strategy

πŸ“– About This Skill


name: pine-script-strategy description: Write, fix, test, and backtest Pine Script v5 strategies and indicators for TradingView. Use when asked to create trading strategies, indicators, Pine Script code, fix compilation errors, backtest ideas, optimize strategies, write SMC/scalping/trend-following systems, or anything related to TradingView Pine Script. Triggers on phrases like "pine script", "tradingview strategy", "tradingview indicator", "backtest this", "write a strategy", "fix my pine script", "pine script error", "create indicator", "trading strategy".

Pine Script Strategy Builder

Write production-ready Pine Script v5 strategies and indicators for TradingView.

Workflow

1. Understand the request β€” What instrument, timeframe, style (scalping/intraday/swing)? 2. Choose type β€” strategy() for backtesting + signals, indicator() for visual-only 3. Follow architecture β€” Data β†’ Signal β†’ Score β†’ Filter β†’ Risk β†’ Execute β†’ Visual 4. Apply non-repaint rules β€” MANDATORY, no exceptions 5. Run pre-flight checklist β€” Before finalizing ANY script 6. Test mentally β€” Walk through edge cases, verify logic on paper 7. Deliver β€” Complete script + explanation + suggested backtest settings

Non-Negotiable Rules

Non-Repaint (ALWAYS)

  • Wrap signals in barstate.isconfirmed
  • strategy() must have calc_on_every_tick=false, process_orders_on_close=false
  • request.security() β†’ use barmerge.gaps_off, barmerge.lookahead_on with [1] offset
  • NEVER use current bar data for signal generation without confirmation
  • Risk Management (ALWAYS)

  • Every trade MUST have a stop loss β€” no naked positions
  • Use ATR-based or swing-based SL
  • RR ratio β‰₯ 1:2 preferred
  • Max trades per day filter
  • Cool-off after loss
  • Quality Gates (ALWAYS)

  • ADX chop filter (ADX > 20 to trade)
  • Session filter for appropriate trading hours
  • Volume confirmation (1.2x+ average)
  • Score threshold β‰₯ 70 for entries
  • Strategy Architecture

    Layer 1: Data β€” Collect indicators (EMA, RSI, ATR, VWAP, ADX, Volume, BB, Pivots)
    Layer 2: Signal β€” Generate signals (trend direction, breakouts, pullbacks, sweeps)
    Layer 3: Score/Filter β€” Score + filter (multi-factor score, chop filter, session, cool-off)
    Layer 4: Risk β€” SL/TP calculation (ATR/swing-based, partial exits, trailing)
    Layer 5: Execute β€” Entry/exit logic (NON-REPAINT, barstate.isconfirmed)
    Layer 6: Visual β€” Labels, boxes, tables, alerts, backgrounds
    

    Pre-Flight Checklist

    Before delivering ANY strategy, verify:

  • [ ] Non-repaint logic (barstate.isconfirmed)?
  • [ ] Chop filter (ADX > 20)?
  • [ ] Session filter?
  • [ ] Cool-off after loss?
  • [ ] Score threshold β‰₯ 70?
  • [ ] Strong breakout confirmation?
  • [ ] Full EMA alignment (3+ EMAs)?
  • [ ] Volume confirmation (1.2x+)?
  • [ ] RR β‰₯ 1:2?
  • [ ] Max trades per day?
  • [ ] SL always set?
  • [ ] Partial exit or trailing?
  • Templates

    Strategy Template

    //@version=5
    strategy("Name", overlay=true,
         initial_capital=10000,
         default_qty_type=strategy.percent_of_equity,
         default_qty_value=10,
         commission_type=strategy.commission.percent,
         commission_value=0.05,
         slippage=1,
         pyramiding=0,
         process_orders_on_close=false,
         calc_on_every_tick=false)
    

    Indicator Template

    //@version=5
    indicator("Name", overlay=true, max_lines_count=500, max_labels_count=500)
    

    Key Patterns

    Non-Repaint Entry

    var bool longSignal = false
    if barstate.isconfirmed
        longSignal := 
    else
        longSignal := false

    if longSignal and strategy.position_size == 0 and barstate.isconfirmed strategy.entry("Long", strategy.long)

    Multi-Timeframe Non-Repaint

    htfClose = request.security(syminfo.tickerid, "60", close[1], barmerge.gaps_off, barmerge.lookahead_on)
    

    ATR-Based SL/TP

    atrVal = ta.atr(14)
    slLong = close - atrVal * 1.5
    tpLong = close + atrVal * 1.5 * rrRatio
    

    Score System (0-100)

    score = 0
    if close > ema50 and close > ema200  // Trend: 0-25
        score += 25
    if rsi > 60                          // Momentum: 0-20
        score += 20
    if volume > ta.sma(volume, 10) * 1.2  // Volume: 0-15
        score += 15
    if close > high[1] and close > open   // Breakout: 0-20
        score += 20
    if ta.atr(14) > ta.atr(14)[1]        // Volatility: 0-10
        score += 10
    if close > ta.vwap(hlc3)              // VWAP: 0-10
        score += 10
    totalScore = math.min(score, 100)
    

    Cool-off After Loss

    var int barsSinceLoss = 999
    if strategy.closedtrades > 0
        if strategy.closedtrades.profit(strategy.closedtrades - 1) < 0
            barsSinceLoss := 0
    barsSinceLoss += 1
    coolOffOK = barsSinceLoss > 5
    

    Common Errors & Fixes

    | Error | Fix | |-------|-----| | Cannot call ta.dmi().adx | [diPlus, diMinus, adxVal] = ta.dmi(len, len) | | Undeclared identifier | Declare with var before use | | Cannot call operator [] on bool | Store in var variable first | | Script could not be translated | Check commas, parentheses, v5 syntax | | Too many plot calls | Max 64 plots β€” use tables/labels | | Loop takes too long | Max 500K iterations β€” reduce loop | | Variable type mismatch | Use float() or int() casting |

    Detailed References

  • SMC patterns (BOS, ChoCH, FVG, Order Blocks, Sweeps): See references/smc-patterns.md
  • All indicators cheat sheet: See references/indicators-cheatsheet.md
  • Entry patterns (proven): See references/entry-patterns.md
  • Risk management patterns: See references/risk-management.md
  • Visual & alert patterns: See references/visuals-alerts.md
  • Performance tracking: See references/performance-tracking.md
  • Output Format

    When delivering a Pine Script: 1. Complete, copy-paste-ready code 2. Brief explanation of what it does 3. Suggested TradingView backtest settings (timeframe, dates, instrument) 4. Known limitations or things to watch for 5. Ideas for improvement

    Fixing Errors

    When fixing Pine Script compilation errors: 1. Read the error message carefully 2. Match to known error patterns above 3. Fix the specific issue β€” don't rewrite the whole script 4. Verify the fix doesn't break other logic 5. Re-check non-repaint compliance

    Optimization Notes

  • Don't overfit to historical data
  • Test across multiple timeframes
  • Verify with walk-forward (test on different date ranges)
  • Keep strategies simple β€” complex β‰  profitable
  • If win rate < 50% or profit factor < 1.3, strategy needs work