Alpha Leak is a production-grade, multi-phase trading intelligence system built for Pump.fun on Solana. It processes every trade, token creation, and graduation event in real time, runs them through layered analysis — wallet scoring, adversarial detection, ML inference, and market regime classification — and executes trades atomically when high-conviction signals emerge.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/alphaleaks60-maker/solvedocs2/llms.txt
Use this file to discover all available pages before exploring further.
Introduction
Understand what Alpha Leak does and how its 30+ concurrent services work together.
Architecture
Full pipeline diagram mapping every service from on-chain ingestion to live execution.
Quickstart
Connect to the live SSE feed and query historical data in under five minutes.
API Reference
REST endpoints and SSE stream reference with full request and response schemas.
How the system works
Alpha Leak runs as a single process with 30+ background services coordinated through a database and cache layer. Events flow through four named phases, each building on the data produced by the previous one.Ingest on-chain events
The
EventProcessor receives every trade, token creation, and graduation event the moment it lands on-chain — no polling, no batching. Events are decoded and fanned out simultaneously to all relevant services.Score wallets and build intelligence
Phase 1 computes a 7-component alpha score for every wallet. Phase 2 classifies token lifecycle state, detects coordinated bundles, scores creator risk, and builds a wallet co-occurrence graph.
Run ML inference
Phase 3 scores signals every 5 seconds using ONNX LightGBM models with Platt-calibrated probabilities. Adversarial detection, copy-trade classification, and market regime run concurrently.
Explore by topic
The Pipeline
Deep dives into ingestion, Phase 1 scoring, Phase 2 intelligence, and Phase 3 ML and advanced detection.
Intelligence
Wallet scoring, adversarial detection, Genesis Watcher, and market regime classification.
ML System
ONNX model deployment, 68-feature vector design, and model training workflows.
Live Trader
Phase-gated strategies, position sizing, circuit breakers, and real-time monitoring.