Aegus PATH
PATH is a memecoin research platform and a multi-asset trading platform. Discover narratives the moment they reawaken. Trade across four asset classes from one hub. Every entry routes through the same deterministic gate stack — no autonomous decisions, no black boxes.
Each class has its own engines tuned to its volatility, liquidity, and narrative cadence. All flow through the same gate stack so comparisons across classes use one shape of evidence.
Narrative-reawakening detection across the universe. Active-mover bridge, dormant-volume-spike, and equity-base sources gate entries on smart-money cohort + narrative tailwind.
RSI + EMA crossover strategies on 4h candles. Position sizing scaled by market regime (risk-on / risk-off).
PAXG / XAUT spread arbitrage and DCA cycles. Macro inflation-hedge correlation overlay.
Tokenized AAPLX / NVDAX / MSTRX / CRCLX / GOOGLX bases. Equity-base detector emits narrative-aligned entries with longer hold cadence (24h vs 12h NA cap).
No autonomous AI deciding trades. The LLM appears once as an advisory grading gate. Everything else is rule-based and reversible.
22 signal engines surface candidate tokens — gmgn trending, KOL flows, dormancy reawakenings, narrative momentum, on-chain whale flows.
Each candidate enriched with rug-check scores, holder distribution, liquidity verification, narrative tags, and OHLCV-derived ATR.
Rule-based gates (RugCheck, SolSniffer, smart-money cohort, source allowlist, asset-class routing, deep-research LLM verdict) reject anything that doesn't clear.
Survivors get position-sized per source preset and asset class. Stop-loss, take-profit, and ATR-adaptive trail set at entry. Paper by default.
Exit loop fires on TP / SL / trail-drop / time-based stale. Every close logged with realized PnL, gate-trace, and outcome attribution.
Institute grades every signal post-hoc. Sources with negative paper edge get throttled. Engines with positive edge get promoted toward live.
We measure paper-edge before any capital moves. Live trading is gated on a measurable refund-criteria bar — closes, weighted edge, win rate, and live-eligibility — sustained across a meaningful cohort. We publish our progress on GitHub.