OPEN-SOURCE SCRIPT
AetherEdge - Mini Multi-Agent Consensus
🖊️ Overview
AE-MMAC is a consensus signal where three independent agents — Trend, Momentum, Volume — each vote BUY/SELL/wait, and a single clear arrow prints only when a majority agrees. Every agent's verdict is shown on a dashboard, letting you trade with the "leave it to the AI" feel without losing transparency. Agents' past performance is online-learned so high-reliability agents speak louder in the consensus.
🔶 Key Features
3 independent agents voting: Trend (EMA), Momentum (RSI), Volume (z-score)
Majority consensus: fires only when 2+ agents agree; otherwise waits
Weighted voting: each agent's recent accuracy adjusts its vote weight
One-arrow design: a single BUY/SELL label on consensus flips — no clutter
Transparent dashboard: each agent's vote, weight, and recent accuracy at a glance
Optional candle coloring
Alerts on consensus signals
🧠 Technical Architecture
This tool is not a true multi-agent reinforcement-learning system but combines three independent classical rules with reliability learning. The trend agent compares short vs long EMAs, the momentum agent thresholds RSI, and the volume agent votes on the candle direction when volume z-score exceeds a threshold. Each new agent vote is buffered; after the outcome horizon, a label is generated from whether an ATR-scaled move occurred in the vote direction, updating the weight via W ← W + α(won − W). Consensus checks both the raw vote count (minimum votes) and the sign of the weighted sum, so low-reliability agents cannot distort the verdict. Signals fire only as an edge trigger on consensus flips, preventing repetition.
⚙️ Recommended Settings & Tuning Guide
Tuned with crypto in mind. Defaults (Trend EMA 10/30, RSI 14 with 55/45 thresholds, Volume z 1.0) are a starting point. Min Votes = 2 (majority) is recommended; 3 (unanimous) tends to be too strict. Keep Use Learned Agent Weights on so the consensus grows smarter with experience. On high-volatility instruments, raise Volume z Threshold to 1.5 to reduce noise.
💡 How to Use in Practice
When an arrow fires, check the dashboard first. A 3-0 vote is the strongest; 2-1 means majority with dissent. Look at per-agent accuracy too — if the trend agent is at 70% and the others at 50%, the market is trend-led. Scalpers can enter immediately, day traders can wait for the first pullback. Pair with structure tools and prioritize moments where the consensus aligns with key levels.
⚠️ Important Notes
This tool is multi-agent-style, not a true multi-agent RL system. The agents are independent classical rules without inter-agent communication; the AI element is limited to learning each agent's reliability. Learning needs accumulated data, so on new instruments or short windows the agent weights stay near the initial 0.5 and the consensus collapses to simple majority. Empty or near-50% accuracy cells in the HUD signal insufficient learning data.
🚨 Disclaimer
This indicator is provided for educational and informational purposes only and does not constitute financial or investment advice. No signal guarantees future profit, and past performance does not indicate future results. All trading decisions are made at your own risk.
AE-MMAC is a consensus signal where three independent agents — Trend, Momentum, Volume — each vote BUY/SELL/wait, and a single clear arrow prints only when a majority agrees. Every agent's verdict is shown on a dashboard, letting you trade with the "leave it to the AI" feel without losing transparency. Agents' past performance is online-learned so high-reliability agents speak louder in the consensus.
🔶 Key Features
3 independent agents voting: Trend (EMA), Momentum (RSI), Volume (z-score)
Majority consensus: fires only when 2+ agents agree; otherwise waits
Weighted voting: each agent's recent accuracy adjusts its vote weight
One-arrow design: a single BUY/SELL label on consensus flips — no clutter
Transparent dashboard: each agent's vote, weight, and recent accuracy at a glance
Optional candle coloring
Alerts on consensus signals
🧠 Technical Architecture
This tool is not a true multi-agent reinforcement-learning system but combines three independent classical rules with reliability learning. The trend agent compares short vs long EMAs, the momentum agent thresholds RSI, and the volume agent votes on the candle direction when volume z-score exceeds a threshold. Each new agent vote is buffered; after the outcome horizon, a label is generated from whether an ATR-scaled move occurred in the vote direction, updating the weight via W ← W + α(won − W). Consensus checks both the raw vote count (minimum votes) and the sign of the weighted sum, so low-reliability agents cannot distort the verdict. Signals fire only as an edge trigger on consensus flips, preventing repetition.
⚙️ Recommended Settings & Tuning Guide
Tuned with crypto in mind. Defaults (Trend EMA 10/30, RSI 14 with 55/45 thresholds, Volume z 1.0) are a starting point. Min Votes = 2 (majority) is recommended; 3 (unanimous) tends to be too strict. Keep Use Learned Agent Weights on so the consensus grows smarter with experience. On high-volatility instruments, raise Volume z Threshold to 1.5 to reduce noise.
💡 How to Use in Practice
When an arrow fires, check the dashboard first. A 3-0 vote is the strongest; 2-1 means majority with dissent. Look at per-agent accuracy too — if the trend agent is at 70% and the others at 50%, the market is trend-led. Scalpers can enter immediately, day traders can wait for the first pullback. Pair with structure tools and prioritize moments where the consensus aligns with key levels.
⚠️ Important Notes
This tool is multi-agent-style, not a true multi-agent RL system. The agents are independent classical rules without inter-agent communication; the AI element is limited to learning each agent's reliability. Learning needs accumulated data, so on new instruments or short windows the agent weights stay near the initial 0.5 and the consensus collapses to simple majority. Empty or near-50% accuracy cells in the HUD signal insufficient learning data.
🚨 Disclaimer
This indicator is provided for educational and informational purposes only and does not constitute financial or investment advice. No signal guarantees future profit, and past performance does not indicate future results. All trading decisions are made at your own risk.
Script open-source
Dans l'esprit TradingView, le créateur de ce script l'a rendu open source afin que les traders puissent examiner et vérifier ses fonctionnalités. Bravo à l'auteur! Bien que vous puissiez l'utiliser gratuitement, n'oubliez pas que la republication du code est soumise à nos Règles.
Clause de non-responsabilité
Les informations et publications ne sont pas destinées à être, et ne constituent pas, des conseils ou recommandations financiers, d'investissement, de trading ou autres fournis ou approuvés par TradingView. Pour en savoir plus, consultez les Conditions d'utilisation.
Script open-source
Dans l'esprit TradingView, le créateur de ce script l'a rendu open source afin que les traders puissent examiner et vérifier ses fonctionnalités. Bravo à l'auteur! Bien que vous puissiez l'utiliser gratuitement, n'oubliez pas que la republication du code est soumise à nos Règles.
Clause de non-responsabilité
Les informations et publications ne sont pas destinées à être, et ne constituent pas, des conseils ou recommandations financiers, d'investissement, de trading ou autres fournis ou approuvés par TradingView. Pour en savoir plus, consultez les Conditions d'utilisation.