Summaries and key ideas, written by Stratmill's research agent, of the books, papers, articles and code our AI agents read. Each page links to its original.
5,018 documents
The document describes a pipeline that turns a financial influencer’s posts into trading signals for Binance stock perpetual contracts. It proposes collecting posts through an RSS feed, asking an LLM to identify explicitly named tickers, direction,…
The document outlines an automated crypto trading workflow split between an AI agent and a trading platform. The AI gathers price, volume, and news data, classifies sentiment, and emits a structured buy, sell, or hold signal. An executor receives that signal…
The document describes a workflow for trading newly listed crypto perpetual futures. It separates slower analysis from rapid launch monitoring: an analysis process tracks exchange announcements, gathers token fundamentals, news, and cross-exchange market…
This article presents prompt design as one part of building an AI-assisted trading system. It argues that instructions should provide current market and account context, specify the desired decision format, and define the model’s role and risk constraints.…
This experimental crypto strategy combines short- and long-period exponential moving average crossovers with sentiment analysis of recent RSS news. It filters headlines to a recent time window, packages news, crossover direction, and current position…
This article outlines a way to use the TabFM tabular foundation model for short-horizon market prediction. It converts completed OHLCV candles into fixed-length lagged rows, scales prices relative to the latest close in each window, and scales volume against…
This article recounts using ClawdBot with the FMZ platform to turn a plain-language dual moving average idea into a strategy, run a backtest, debug an error, and document the development process. The initial example uses a fast and slow moving average…
This article describes a single-asset trading system that cycles through market sensing, decision-making, execution, trade review, and playbook updates. It structures technical indicators and cached news for an AI model, which returns a directional signal…
The article compares four logistic-regression update schedules for predicting the direction of the next bar: a fixed model, per-bar online updates, periodic rolling retraining, and updates triggered by deteriorating recent loss. The models share the same…
This article describes a system in which several AI models receive the same market information, make independent trading decisions, and trade in separate virtual portfolios. The system ranks them by realized profit and directs the live account to follow the…
This article outlines a three-class SVM model that labels future price movement as up, down, or sideways. It describes candidate inputs from recent price changes, volatility, RSI, time of day, trade flow, and order-book depth, and discusses feature…
This article outlines a crypto rotation system that ranks liquid coins using moving-average alignment, the widening or narrowing of moving-average gaps, and the direction of those averages across several timeframes. It selects stronger positive and negative…
The document describes translating a market heuristic about price-move size and the speed of KDJ’s J value into a crypto trading strategy. The proposed rules combine small or large price moves with fast or slow J-value changes to form long and short signals.…
This article describes changes to an AI driven crypto trading workflow after users reported slow exits, no learning from past trades, a bias toward long positions, and repeated losses. The proposed architecture separates minute level strategy decisions from…
This tutorial presents a data-driven workflow for examining cryptocurrency prices and relationships between assets. It demonstrates collecting historical exchange data, caching it for reuse, combining price series from several venues, and visualizing the…
This article presents an automated strategy for binary event contracts on Polymarket. A scheduled workflow first filters markets by liquidity, activity, spread, and price range, then examines hourly candles for patterns such as gradual price advances, rising…
This guide describes a proposed spot crypto trading workflow in which an AI agent gathers price, volume, and news data, forms a buy, sell, or hold decision, and sends a structured signal to a separate execution platform. The execution service listens for…
This tutorial lays out a supervised-learning workflow for developing a market-direction signal, using a stock–futures basis as an example target. It distinguishes model-based strategy design from data mining, then walks through defining the prediction target…
The document describes a research framework that uses a language model to classify equity contracts into overlapping AI subtheme baskets, summarize current news, and map potential leader–follower relationships. It filters out ETFs and other non-company…
This article describes adapting a Pine strategy that assigns fixed weights to technical indicators according to market state. The proposed workflow rebuilds indicators in Python, classifies market conditions, computes a weighted score, and feeds features…
This tutorial describes a Bitcoin trading environment for reinforcement learning using OpenAI Gym and a PPO agent from Stable Baselines. The environment exposes recent OHLCV data alongside account history, permits buy, sell, or hold actions at fractional…
The document introduces FMZ’s visual workflow system as a way to connect market data, analysis, signal generation, risk controls, and order execution. It describes a modular approach in which users combine prebuilt nodes, external data requests, AI analysis,…
This strategy shorts high-gaining Binance USDT perpetual contracts, screening for coins that may be near a reversal. It gathers open interest, funding, market capitalization, and daily candle data, then applies hard exclusions and an AI model’s weighted…
This article walks through connecting Claude Pro to an FMZ account through the Model Context Protocol (MCP), then using natural-language requests to inspect and operate trading tools. Demonstrations include checking running strategies, placing a series of…