Dvir Ben Ishay
HEBook a call
Book an intro call
Detail 01 · Open-source capital markets · DBI-M · rev.26

Open-source tools for capital markets

Open-source libraries, bots and platforms for algorithmic trading, backtesting, market data, portfolios and crypto, including AI agents that analyze stocks and suggest decisions. Next to them, what developers and researchers write about building such tools. Collected every morning from GitHub and open sources.

Updated daily. Information only, not investment advice.

Updated
Sources
GitHub search API, Hacker News, DEV, arXiv, Medium
Tools
69
Articles
72

Read this before you use anything here

This page is for information only. It is not investment advice, not investment marketing and not a recommendation to buy or sell any security, currency or asset. Dvir Ben Ishay is not a licensed investment advisor.

The tools here are third-party open-source code with no warranty. A trading tool can have bugs, place orders you did not intend, lose real money or leak your account's API keys. Check every tool yourself, start on a demo account, and talk to a licensed professional before you put real money on it.

Detail 02 · The tools

Open-source tools by field

In each field, the projects with the most GitHub stars. Every tool has its own page with its description, language and last update.

Israel and TASE

9 tools

Tools for the Tel Aviv Stock Exchange and the Israeli market: TASE data, scanners and portfolio trackers. The field is small, so projects with few stars are included.

  1. Rank 1

    asafravid/sss

    161 stars

    Stock Scanner & Screener: A yfinance-based Stock Scanner and Screener, focusing on Fundamental Properties of scanned stocks.

    Language:
    Python
    Opened:
    Last update:
    • nasdaq
    • nasdaq100
    • python
    • python3
    • russel1000
    • snp500
  2. Rank 2

    algonell/tase

    9 stars

    Tel Aviv stock exchange data scrapper.

    Language:
    Jupyter Notebook
    Opened:
    Last update:
    • exchange
    • python
    • stocks-trading
  3. Rank 3

    JPDevelop/taseapi

    2 stars

    A python library for the Tel-Aviv Stock Exchange API

    Language:
    Python
    Opened:
    Last update:
  4. Rank 4

    deepvalueinvesting/deep-value-tase-plugin

    2 stars

    Professional financial analysis tools for the Israeli stock market (TASE), Claude Code plugin

    Language:
    Not set
    Opened:
    Last update:
  5. Rank 5

    StockerAPI/israel-stock-market-api

    2 stars

    Real-time Israel stock market API with WebSocket, OHLCV, and snapshot endpoints for TASE.

    Language:
    Not set
    Opened:
    Last update:
    • financial-data
    • il
    • israel-stock-market
    • market-data-api
    • ohlcv-api
    • stock-market-api
  6. Rank 6

    NitzanLary/Takumi

    2 stars

    Personal trading intelligence platform, Israeli (TASE) and US equities via IBI, with Claude-powered AI chat agent

    Language:
    TypeScript
    Opened:
    Last update:
  7. Rank 7

    yonicd/taseR

    1 stars

    read data from tel aviv stock exchange (tase) website to r

    Language:
    HTML
    Opened:
    Last update:
  8. Rank 8

    yotam-sh/trade-vault

    1 stars

    Self-hosted TASE/IBI portfolio tracker, Flask + CLI, multi-portfolio, tax lots, P&L analytics, and a rebalance optimizer.

    Language:
    Python
    Opened:
    Last update:
    • docker
    • fintech
    • flask
    • gunicorn
    • hebrew
    • investing
  9. Rank 9

    benbaron2004/StockMonitoringProject

    1 stars

    TASE stock spread monitoring with configurable threshold alerts and email notifications. Built with claude code

    Language:
    Python
    Opened:
    Last update:

AI trading agents

12 tools

Systems where a language model reads news, filings and market data, and sometimes suggests a decision or places an order. Some are research frameworks, some are ready to run.

  1. Rank 1

    TauricResearch/TradingAgents

    121,906 stars

    TradingAgents: Multi-Agents LLM Financial Trading Framework

    Language:
    Python
    Opened:
    Last update:
    • agent
    • finance
    • llm
    • multiagent
    • trading
  2. Rank 2

    HKUDS/Vibe-Trading

    38,019 stars

    "Vibe-Trading: Your Personal Trading Agent"

    Language:
    Python
    Opened:
    Last update:
    • ai-agent
    • algorithmic-trading
    • backtesting
    • fintech
    • llm
    • mcp
  3. Rank 3

    OpenByteInc/QuantDinger

    13,686 stars

    The most complete open-source, self-hosted Agentic Trading OS, AI research, Python strategies, factor discovery, backtesting, optimization, paper/live multi-asset execution, risk controls, monitoring, MCP, and multi-tenant SaaS in one auditable stack.

    Language:
    Python
    Opened:
    Last update:
    • agentic-ai
    • ai-agents
    • backtesting
    • crypto
    • finance
    • fintech
  4. Rank 4

    NoFxAiOS/nofx

    13,254 stars

    Your AI trading terminal assistant for US stocks, commodities, forex, and crypto.

    Language:
    Go
    Opened:
    Last update:
    • agentictrading
    • ai
    • ai-trading
    • arbitrage
    • binance
    • bitcoin
  5. Rank 5

    Drakkar-Software/OctoBot

    6,922 stars

    Free open source crypto trading bot to automate AI, Grid, DCA and TradingView strategies on Binance, Hyperliquid and 15+ exchanges, with a simple interface.

    Language:
    Python
    Opened:
    Last update:
    • ai-trading
    • backtest
    • binance-api
    • bitcoin
    • coinbase-bot
    • cryptocurrencies
  6. Rank 6

    simonlin1212/TradingAgents-astock

    3,945 stars

    A-share multi-agent investment research framework, adapted to A-share data sources (Dragon Tiger List/hot money/unlocking, etc.), 7 analysts' debate and decision-making based on A-share rules, deeply modified based on TradingAgents, adapted to the A-share market. A-share multi-agent investment research framework, 7 AI analysts, bull/bear debate, risk assessment.

    Language:
    Python
    Opened:
    Last update:
    • a-share
    • ai-agent
    • china-stocks
    • claude
    • fintech
    • investment-research
  7. Rank 7

    wquguru/nof0

    2,753 stars

    Open-source AI trading arena. A full reproduction of NOF1 Alpha Arena for running and comparing LLM agents.

    Language:
    Go
    Opened:
    Last update:
    • agentic-ai
    • ai
    • ai-agents
    • crypto
    • golang
    • hyperliquid
  8. Rank 8

    shirosaidev/stocksight

    2,552 stars

    Stock market analyzer and predictor using Elasticsearch, Twitter, News headlines and Python natural language processing and sentiment analysis

    Language:
    Python
    Opened:
    Last update:
    • elasticsearch
    • natural-language-processing
    • nltk
    • python
    • sentiment
    • sentiment-analysis
  9. Rank 9

    TNT-Likely/PanWatch

    2,445 stars

    PanWatch, AI stock monitoring for A-shares, HK & US markets, powered by TradingAgents. Portfolio insights, real-time alerts & automated reports. | PanWatch: AI stock monitoring, portfolio analysis, real-time alerts and automated reports covering A-shares, HK and US stocks.

    Language:
    Python
    Opened:
    Last update:
    • a-share
    • agent
    • ai
    • ai-agent
    • akshare
    • deepseek
  10. Rank 10

    achillesrasquinha/bulbea

    2,347 stars

    Deep Learning based Python Library for Stock Market Prediction and Modelling

    Language:
    Python
    Opened:
    Last update:
    • deep-learning
    • finance
    • machine-learning
    • prediction
    • python-library
    • quantitative-finance
  11. Rank 11

    ginlix-ai/LangAlpha

    2,268 stars

    Claude Code for Financial Market

    Language:
    Python
    Opened:
    Last update:
    • agent
    • investment
    • langchain
    • langraph
    • llm
    • mcp
  12. Rank 12

    Lumiwealth/lumibot

    2,224 stars

    Python trading framework with AI agents that actually place the trade. Backtest, paper and live trade on 12 brokers: stocks, options, futures, forex, crypto and prediction markets.

    Language:
    Python
    Opened:
    Last update:
    • agentic-trading
    • ai-agents
    • ai-trading
    • algorithmic-trading
    • alpaca
    • backtesting

Crypto bots and exchanges

12 tools

Crypto trading bots and libraries that connect to crypto exchanges through one API.

  1. Rank 1

    freqtrade/freqtrade

    56,751 stars

    Free, open source crypto trading bot

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • bitcoin
    • cryptocurrencies
    • cryptocurrency
    • freqtrade
    • python
  2. Rank 2

    ccxt/ccxt

    44,856 stars

    A unified trading API with more than 100 crypto exchanges and prediction markets in JavaScript / TypeScript / Python / C# / PHP / Go / Java / Rust

    Language:
    Python
    Opened:
    Last update:
    • altcoin
    • api
    • arbitrage
    • bitcoin
    • bot
    • btc
  3. Rank 3

    hummingbot/hummingbot

    21,153 stars

    Open source software that helps you create and deploy high-frequency crypto trading bots

    Language:
    Python
    Opened:
    Last update:
    • ai-agents
    • algotrading
    • arbitrage
    • backtesting
    • binance
    • bitcoin
  4. Rank 4

    jesse-ai/jesse

    8,795 stars

    An advanced crypto trading bot written in Python

    Language:
    Python
    Opened:
    Last update:
    • algo-trading
    • algorithmic-trading
    • bitcoin
    • bot
    • crypto
    • crypto-bot
  5. Rank 5

    Superalgos/Superalgos

    5,784 stars

    Free, open-source crypto trading bot, automated bitcoin / cryptocurrency trading software, algorithmic trading bots. Visually design your crypto trading bot, leveraging an integrated charting system, data-mining, backtesting, paper trading, and multi-server crypto bot deployments.

    Language:
    JavaScript
    Opened:
    Last update:
    • algorithmic-trading
    • algotrading
    • bitcoin-trading
    • bitcoin-trading-bots
    • crypto
    • crypto-trading
  6. Rank 6

    CryptoSignal/Crypto-Signal

    5,659 stars

    Github.com/CryptoSignal, Trading & Technical Analysis Bot, 4,100+ stars, 1,100+ forks

    Language:
    Python
    Opened:
    Last update:
    • abenezer-mamo
    • algorithmic-trading
    • binance
    • binance-api
    • bitcoin
    • bittrex-api
  7. Rank 7

    nkaz001/hftbacktest

    5,071 stars

    Free, open source, a high frequency trading and market making backtesting and trading bot, which accounts for limit orders, queue positions, and latencies, utilizing full tick data for trades and order books(Level-2 and Level-3), with real-world crypto trading examples for Binance and Bybit

    Language:
    Rust
    Opened:
    Last update:
    • algorithmic-trading
    • algotrading
    • backtesting
    • backtesting-engine
    • backtesting-trading-strategies
    • binance
  8. Rank 8

    ctubio/Krypto-trading-bot

    3,735 stars

    Self-hosted crypto trading bot (automated high frequency market making) written in C++

    Language:
    C++
    Opened:
    Last update:
    • binance
    • bitcoin
    • bitfinex
    • bitmex
    • bob-marley
    • bot-platform
  9. Rank 9

    Haehnchen/crypto-trading-bot

    3,551 stars

    Cryptocurrency trading bot in javascript for Bitfinex, Bitmex, Binance, Bybit ... (public edition)

    Language:
    TypeScript
    Opened:
    Last update:
    • binance
    • bitfinex
    • bitmex
    • bot
    • bybit
    • ccxt
  10. Rank 10

    iterativv/NostalgiaForInfinity

    3,527 stars

    Trading strategy for the Freqtrade crypto bot

    Language:
    Python
    Opened:
    Last update:
  11. Rank 11

    alsk1992/CloddsBot

    3,098 stars

    Open Source AI trading agent that operates autonomously across 1000+ markets, Polymarket, Kalshi, Binance, Hyperliquid, Solana DEXs, 5 EVM chains. Scans for edge, executes instantly, manages risk while you sleep. Agent commerce protocol for machine-to-machine payments. Self-hosted. Built on Claude.

    Language:
    TypeScript
    Opened:
    Last update:
    • agi
    • ai
    • arbitrage
    • claude
    • crypto
    • defi
  12. Rank 12

    Open-Trader/opentrader

    2,925 stars

    Open-source crypto trading bot | DCA & GRID strategies | UI | Star to support the project!

    Language:
    TypeScript
    Opened:
    Last update:
    • algotrading
    • bot
    • crypto
    • cryptocurrency
    • custom-strategy
    • dca-bot

Market data

12 tools

Libraries and services for prices, trading data and financial data, for research, testing and bots.

  1. Rank 1

    microsoft/qlib

    50,205 stars

    Qlib is an AI-oriented Quant investment platform that aims to use AI tech to empower Quant Research, from exploring ideas to implementing productions. Qlib supports diverse ML modeling paradigms, including supervised learning, market dynamics modeling, and RL, and is now equipped with https://github.com/microsoft/RD-Agent to automate R&D process.

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • auto-quant
    • deep-learning
    • finance
    • fintech
    • investment
  2. Rank 2

    wilsonfreitas/awesome-quant

    30,572 stars

    A curated list of insanely awesome libraries, packages and resources for Quants (Quantitative Finance)

    Language:
    HTML
    Opened:
    Last update:
    • algorithmic-trading-engine
    • algorithmic-trading-library
    • algotrading
    • arbitrage-bot
    • awesome
    • awesome-list
  3. Rank 3

    ranaroussi/yfinance

    25,887 stars

    Download market data from Yahoo! Finance's API

    Language:
    Python
    Opened:
    Last update:
    • financial-data
    • fix-yahoo-finance
    • market-data
    • pandas
    • python
    • stock-data
  4. Rank 4

    akfamily/akshare

    23,288 stars

    AKShare is an elegant and simple financial data interface library for Python, built for human beings! Open source financial data interface library

    Language:
    Python
    Opened:
    Last update:
    • academic
    • akshare
    • asset-pricing
    • bond
    • currency
    • data
  5. Rank 5

    questdb/questdb

    17,521 stars

    QuestDB is a high performance, open-source, time-series database

    Language:
    Java
    Opened:
    Last update:
    • apache-arrow
    • capital-markets
    • database
    • historian
    • kdb
    • low-latency
  6. Rank 6

    waditu/tushare

    15,489 stars

    TuShare is a utility for crawling historical data of China stocks

    Language:
    Python
    Opened:
    Last update:
    • finance
    • fintech
    • pandas
    • python
    • realtime-quotes-data
    • stock-data
  7. Rank 7

    simonlin1212/a-stock-data

    11,701 stars

    A-share full-stack data toolkit: market K-lines, daily ticks, research reports, signals, fund flow, news, financials, filings, limit-up, ETF options, public sentiment, macro interest rates, futures commodities (including DCE daily K-lines), event-driven, convertible bonds | 15 layers, 87 endpoints, 34 data sources, key-free except for iwencai | A-share data for AI agents: K-lines, ticks, reports, fund flow, news, financials, filings, options, macro, futures, events, convertibles | 15 layers, 87 endpoints, 34 sources

    Language:
    Python
    Opened:
    Last update:
    • a-share
    • ai-agent
    • ai-skill
    • china-stocks
    • claude-code
    • financial-data
  8. Rank 8

    JerBouma/FinanceDatabase

    9,610 stars

    This is a database of 300.000+ symbols containing Equities, ETFs, Funds, Indices, Currencies, Cryptocurrencies and Money Markets.

    Language:
    Python
    Opened:
    Last update:
    • analysis
    • cryptocurrencies
    • currencies
    • database
    • equities
    • etfs
  9. Rank 9

    JerBouma/FinanceToolkit

    5,517 stars

    Transparent and Efficient Financial Analysis

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • equities
    • factor-analysis
    • finance
    • financial-analysis
    • financial-data
  10. Rank 10

    1nchaos/adata

    5,385 stars

    Free and open-source A-share quantitative trading database; focusing on A-shares, focusing on quant, growing towards the sun; open, pure, continuous, generating power for AI (love). Born for individual quantitative trading, defending 3000 points, cherishing bottom opportunities... [stock data, stock market data, stock quantitative data, stock trading data, K-line market data, stock concept data, stock data interface, market data interface, quantitative trading data] [multi-source data fusion, dynamic proxy settings, ensuring high data availability]

    Language:
    Python
    Opened:
    Last update:
    • 3000
    • a
    • china
    • datasource
    • financial
    • kline
  11. Rank 11

    atilaahmettaner/tradingview-mcp

    5,262 stars

    MCP server for market analysis in AI assistants: screeners, multi-timeframe technical analysis, backtests and sentiment for crypto, stocks, forex and futures. 39 tools. Self-host free, or use the hosted version, TradingMCP.

    Language:
    Python
    Opened:
    Last update:
    • claude-desktop
    • cryptocurrency
    • futures
    • market-data
    • mcp-server
    • mcp-tools
  12. Rank 12

    RomelTorres/alpha_vantage

    4,943 stars

    A python wrapper for Alpha Vantage API for financial data.

    Language:
    Python
    Opened:
    Last update:
    • alpha-vantage
    • alphavantage
    • api-wrapper
    • bitcoin
    • cryptocurrency
    • finance

Portfolio and analysis

12 tools

Portfolio construction, optimization, risk and investment tracking.

  1. Rank 1

    xbtlin/ai-berkshire

    17,086 stars

    Berkshire in the AI era: a value investing research framework based on Claude Code / Codex. Methodologies of four masters: Buffett, Munger, Duan Yongping, and Li Lu + multi-agent parallel research. | AI-era Berkshire: a value investing research framework built for Claude Code / Codex. 4 masters' methodologies + multi-agent adversarial analysis.

    Language:
    HTML
    Opened:
    Last update:
    • ai
    • ai-agent
    • anthropic
    • berkshire-hathaway
    • charlie-munger
    • china-stock
  2. Rank 2

    polakowo/vectorbt

    9,805 stars

    The research engine that gives you an unfair advantage. Run thousands of trading ideas before others finish one.

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • algorithmic-traiding
    • backtesting
    • cryptocurrency
    • data-science
    • data-visualization
  3. Rank 3

    wealthfolio/wealthfolio

    9,412 stars

    A beautiful, private, local-first personal finance tracker. Investments, net worth, spending, and simulations.

    Language:
    Rust
    Opened:
    Last update:
    • macos-app
    • portfolio-tracker
    • self-hosted
    • tauri-app
  4. Rank 4

    PyPortfolio/PyPortfolioOpt

    6,159 stars

    Financial portfolio optimization in python, including classical efficient frontier, Black-Litterman, Hierarchical Risk Parity

    Language:
    Jupyter Notebook
    Opened:
    Last update:
    • algorithmic-trading
    • covariance
    • efficient-frontier
    • finance
    • financial-analysis
    • investing
  5. Rank 5

    hudson-and-thames/mlfinlab

    4,972 stars

    MlFinLab helps portfolio managers and traders who want to leverage the power of machine learning by providing reproducible, interpretable, and easy to use tools.

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • finance
    • financial-machine-learning
    • investing
    • machine-learning
    • portfolio-management
  6. Rank 6

    dcajasn/Riskfolio-Lib

    4,602 stars

    Portfolio Optimization in Python

    Language:
    C++
    Opened:
    Last update:
    • asset-allocation
    • convex-optimization
    • cvar-optimization
    • cvxpy
    • drawdown-model
    • duration-matching
  7. Rank 7

    EliteQuant/EliteQuant

    4,271 stars

    A list of online resources for quantitative modeling, trading, portfolio management

    Language:
    Not set
    Opened:
    Last update:
    • algorithmic-trading
    • asset-management
    • asset-pricing
    • machine-learning
    • mathematical-finance
    • portfolio-management
  8. Rank 8

    rotki/rotki

    4,089 stars

    A portfolio tracking, analytics, accounting and management application that protects your privacy

    Language:
    Python
    Opened:
    Last update:
    • accounting
    • analytics
    • bitcoin
    • blockchain
    • cryptocurrencies
    • cryptocurrency-exchanges
  9. Rank 9

    tradytics/eiten

    3,329 stars

    Statistical and Algorithmic Investing Strategies for Everyone

    Language:
    Python
    Opened:
    Last update:
    • ai
    • algorithmic-trading
    • eigenvalues
    • free-software
    • genetic-algorithm
    • hedgefund
  10. Rank 10

    letianzj/QuantResearch

    3,061 stars

    Quantitative analysis, strategies and backtests

    Language:
    Jupyter Notebook
    Opened:
    Last update:
    • algorithmic-trading
    • algotrading
    • asset-allocation
    • asset-management
    • backtesting-trading-strategies
    • backtests
  11. Rank 11

    gloom-sh/gloomberb

    2,700 stars

    Finance terminal, in your terminal.

    Language:
    TypeScript
    Opened:
    Last update:
    • bloomberg-terminal
    • finance
    • market-data
    • open-source-finance
    • portfolio-tracker
    • stocks
  12. Rank 12

    skfolio/skfolio

    2,620 stars

    Python library for portfolio optimization, factor model construction, and risk management built on scikit-learn

    Language:
    Python
    Opened:
    Last update:
    • asset-allocation
    • backtesting
    • factor-investing
    • factor-models
    • investing
    • investment-analysis

Backtesting and quant libraries

12 tools

Backtesting engines, quant research libraries and algorithmic trading frameworks, plus curated lists.

  1. Rank 1

    openbq-org/OpenBB

    75,313 stars

    Open Data Platform for analysts, quants and AI agents.

    Language:
    Python
    Opened:
    Last update:
    • ai
    • crypto
    • derivatives
    • economics
    • equity
    • finance
  2. Rank 2

    ZhuLinsen/daily_stock_analysis

    69,613 stars

    LLM-powered multi-market stock analysis system with multi-source market data, real-time news, decision dashboard, automated notifications, and cost-free scheduled runs.

    Language:
    Python
    Opened:
    Last update:
    • a-stock
    • ai-agent
    • aigc
    • llm
    • quant
    • quantitative-finance
  3. Rank 3

    Fincept-Corporation/FinceptTerminal

    33,671 stars

    FinceptTerminal is a modern finance application offering advanced market analytics, investment research, and economic data tools, designed for interactive exploration and data-driven decision-making in a user-friendly environment.

    Language:
    C++
    Opened:
    Last update:
    • ai-agents
    • algorithmic-trading
    • bloomberg-terminal
    • cpp
    • finance
    • financial-markets
  4. Rank 4

    mementum/backtrader

    23,714 stars

    Python Backtesting library for trading strategies

    Language:
    Python
    Opened:
    Last update:
    • backtesting
    • metaclass
    • python
    • trading
  5. Rank 5

    stefan-jansen/machine-learning-for-trading

    21,777 stars

    Code for Machine Learning for Trading, 3rd edition, from data sourcing to live execution.

    Language:
    Jupyter Notebook
    Opened:
    Last update:
    • algorithmic-trading
    • artificial-intelligence
    • backtesting
    • data-science
    • deep-learning
    • finance
  6. Rank 6

    quantopian/zipline

    20,229 stars

    Zipline, a Pythonic Algorithmic Trading Library

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • python
    • quant
    • zipline
  7. Rank 7

    bbfamily/abu

    19,159 stars

    Abu quantitative trading system (stocks, options, futures, bitcoin, machine learning) Python-based open-source quantitative trading and quantitative investment framework

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • bitcoin
    • machine-learning
    • matplotlib
    • numpy
    • pandas
  8. Rank 8

    UFund-Me/Qbot

    18,797 stars

    [updating ...] AI-powered Quantitative Investment Research Platform. online docs: https://ufund-me.github.io/Qbot qbot-mini: https://github.com/Charmve/iQuant

    Language:
    Jupyter Notebook
    Opened:
    Last update:
    • backtest
    • bitcoin
    • blockchain
    • deep-learning
    • fintech
    • funds
  9. Rank 9

    AI4Finance-Foundation/FinRL

    17,041 stars

    FinRL: Financial Reinforcement Learning.

    Language:
    Jupyter Notebook
    Opened:
    Last update:
    • algorithmic-trading
    • deep-reinforcement-learning
    • drl-algorithms
    • drl-framework
    • drl-trading-agents
    • finance
  10. Rank 10

    paperswithbacktest/awesome-systematic-trading

    15,302 stars

    A curated list of awesome libraries, packages, strategies, books, blogs, tutorials for systematic trading.

    Language:
    Python
    Opened:
    Last update:
    • algorithmic-trading
    • algotrading
    • alpha
    • arbitrage-bot
    • awesome
    • awesome-list
  11. Rank 11

    myhhub/stock

    15,096 stars

    stock. Get stock data, calculate stock indicators, chip distribution, identify stock patterns, comprehensive stock selection, stock selection strategies, stock verification backtesting, automatic stock trading, support PC and mobile devices.

    Language:
    Python
    Opened:
    Last update:
    • backtest
    • backtesting
    • broker-trading-platform
    • cyq
    • distribution-of-chips
    • position-cost-distribution
  12. Rank 12

    TA-Lib/ta-lib-python

    12,397 stars

    Python wrapper for TA-Lib (http://ta-lib.org/).

    Language:
    Cython
    Opened:
    Last update:
    • finance
    • pattern-recognition
    • python
    • quantitative-finance
    • ta-lib
    • technical-analysis
Detail 03 · Reading

What developers write about building trading tools

Articles, posts and papers on building trading bots, AI agents and recommendation tools. The headline and summary are ours, and the link goes to the source.
  1. MediumPostby AlgoRover

    Backtesting a popular YouTube EMA trading strategy using AlgoRover across multiple timeframes Read on Medium

    The author recreated an exponential moving average strategy found on YouTube and ran four backtests. The goal was to evaluate which timeframes show potential for further testing. This is useful for traders interested in automated strategy validation.

    Original title: One YouTube Strategy, Four Backtests in AlgoRover

    Read on Medium
  2. Hacker NewsPost

    A personal experiment to explore the challenges of starting quantitative trading Read on Hacker News

    Original title: How Hard Can Quant Trading Be? I Tried It to Find Out

    Read on Hacker NewsHacker News discussion
  3. MediumPostby Daniel Aisen

    Developing artificial intelligence tools tailored for institutional trading environments Read on Medium

    The author discusses their changing perspective on implementing AI within institutional trading systems. The post explores the practical challenges and shifts in mindset required to build these tools. It is relevant for developers working on enterprise financial software.

    Original title: Building AI tools for institutional trading

    Read on Medium
  4. MediumPostby Yasanji Ratnaike

    Building a simulated hedge fund managed by a team of eight AI agents Read on Medium

    The author constructed a trading desk where eight AI agents manage a simulated portfolio of one million dollars. The post details the system architecture and key lessons learned from the experiment. It is useful for developers interested in multi-agent AI systems for finance.

    Original title: I Built a Hedge Fund Run by a Team of AI Agents

    Read on Medium
  5. MediumPostby Walid Koussa

    Building a Python crypto trading bot using ccxt with real-world execution handling Read on Medium

    This guide explains how to develop a cryptocurrency trading bot using Python and the ccxt library. It addresses common post-development issues, such as handling incomplete candle data and managing orders on a testnet. Developers looking to build robust execution systems will find this helpful.

    Original title: How to build a Python crypto trading bot with ccxt: data, signals, orders and a testnet

    Read on Medium
  6. MediumPostby Truong Pham | Moon The Train

    Backtesting the TradingView Hull Suite strategy and analyzing repainting issues Read on Medium

    The author performs a backtest of the Hull Suite strategy on TradingView. The post addresses common configuration settings and analyzes whether the strategy suffers from repainting issues. It is useful for technical traders using Pine Script.

    Original title: TradingView Hull Suite Strategy: Repaint, Settings, Backtest

    Read on Medium
  7. MediumPostby Respect Barrister K

    Using an AI operator to manage daily operations of a trading bot Read on Medium

    The author shares operational logs from running a small trading bot managed by an AI agent. The AI operator quickly identified a critical bug in the auto-restart mechanism that had previously gone unnoticed. This write-up is useful for developers interested in AI-driven system administration and bot operations.

    Original title: I Let an AI Run My Trading Bot’s Daily Ops. Its First Find: Auto-Restart Had Never Worked Once.

    Read on Medium
  8. DEVPostby Josep Oriol Carné

    The Martingale: Mathematical Origins and the Inevitability of Ruin Read on DEV

    Read on DEV
  9. DEVPostby Pavel Kazantsev

    Testing thousands of exit rules in quantitative trading and the validation challenge Read on DEV

    The author evaluated 12,720 exit configurations across 14,399 events in 15 cryptocurrency markets using a simulator. The post discusses why the results failed validation and the challenges of multiple testing in quantitative research. It is useful for quant researchers designing backtests and exit strategies.

    Original title: We Tested 12,720 Exit Rules. The Result Wasn't Validated.

    Read on DEV
  10. arXivPaperby Jacky Chen, Ing-Haw Cheng, John Hull et al.

    Reinforcement learning for options market making and hedging under real market conditions Read on arXiv

    The authors evaluate a reinforcement learning agent trained on historical S&P 500 options data to manage inventory risk. They compare its performance in tail-risk protection against traditional delta hedging and agents trained in simulated environments. This study is valuable for quantitative researchers and options market makers.

    Original title: Risk-Sensitive Reinforcement Learning for an Options Market Maker: An Empirical Study with Real S&P 500 Options

    Read on arXiv
  11. arXivPaperby Yebo Feng, Guangyan Gan, Liangrui Pan et al.

    Analyzing whether LLM trading agents track or follow market reactions to news Read on arXiv

    The authors present a multi-agent trading system designed to evaluate how LLM judgments of web news relate to market movements. They investigate whether these models anticipate market reactions or merely react to delayed news reports. This research is useful for developers building news-based algorithmic trading systems.

    Original title: Track or Follow? A Jev-Compatible Multi-Agent Trading System When Web News Lags the Market

    Read on arXiv
  12. arXivPaperby Peixu Xin, Runchuan Feng, Jiazhen Guo et al.

    Detent enables batch-invariant LLM serving and reinforcement learning rollouts with low latency Read on arXiv

    The authors introduce Detent, a framework that ensures reproducible and batch-invariant LLM inference without sacrificing execution speed. By optimizing kernel selection parameters, it achieves deterministic outputs across different batch configurations. This tool is highly useful for developers working on reinforcement learning rollouts and verifiable LLM deployments.

    Original title: Detent: Batch-Invariant LLM Serving and RL Rollouts at Near-Stock Speed

    Read on arXiv
  13. arXivPaperby Jiashun Wang, Beile Wang, Guangyan Gan et al.

    M3FT framework for multi-agent cryptocurrency perpetual futures trading using micro-macro reasoning Read on arXiv

    The authors propose M3FT, a multi-agent framework designed for trading cryptocurrency perpetual futures. The system combines micro-level numerical data with macro-level market interpretations to optimize multi-asset positions. It is useful for quantitative developers building automated crypto trading systems.

    Original title: M3FT: Micro-Macro-Micro Reasoning for Multi-Agent Cryptocurrency Perpetual Futures Trading

    Read on arXiv
  14. MediumPostby Manpreeth Sai

    Backtesting six years of Polymarket data to evaluate betting on favorites. Read on Medium

    The author built a systematic algorithm to analyze historical data from the prediction market Polymarket. The analysis suggests that consistently backing the favorite does not yield profitable results. This is useful for prediction market traders and algorithmic bettors.

    Original title: I backtested 6 Years of Polymarket Data. Backing the Favourite Doesn’t Pay.

    Read on Medium
  15. MediumPostby Kevin Meneses González

    Comparing Backtrader, vectorbt, and Backtesting.py on a ten-year historical dataset. Read on Medium

    The author executed the same trading strategy across three Python backtesting libraries using ten years of data. The results show that Backtrader and vectorbt produced identical outcomes, while Backtesting.py had a minor discrepancy. This comparison is useful for quantitative developers choosing a backtesting framework.

    Original title: Backtrader vs vectorbt vs Backtesting.py: I Ran the Same Strategy on 10 Years of Data

    Read on Medium
  16. MediumPostby Kryptera

    Backtesting MACD trading rules on forty-three stocks from the NIFTY fifty index. Read on Medium

    The author tested different MACD indicator strategies on historical stock data from the Indian market. The results showed that high win-rate rules barely generated profits, while a rule with a high loss rate performed best. This post is useful for technical analysts and retail algorithmic traders.

    Original title: The Indicator Everyone Trusts Lost the Race: Backtesting MACD on 43 NIFTY 50 Stocks

    Read on Medium
  17. arXivPaperby Mark Oskin

    Retrofitting unbounded context windows onto existing language models without retraining Read on arXiv

    The authors demonstrate a method to enable unbounded context windows in stock language models without modifying their weights. By manipulating the geometry of rotary attention, the models can retrieve information from extremely long sequences. This technique is highly relevant for developers building LLM applications that process massive financial documents.

    Original title: Retrofitting Unbounded Context onto Stock Language Models

    Read on arXiv
  18. arXivPaperby Xiangjing An, Zhibin Jiao, Aiting Chen

    StockASkill: A local-first quantitative coprocessor for autonomous stock selection agents Read on arXiv

    The authors present StockASkill, a local-first system designed to reduce latency and API dependency in quantitative stock selection. It integrates a DuckDB database engine containing historical price records and financial disclosures across multiple global markets. This tool is useful for quantitative researchers and developers building autonomous trading agents.

    Original title: StockASkill: A Local-First AgentSkill for Intelligent Stock Selection

    Read on arXiv
  19. MediumPostby BenjaminCup

    Building and deploying a Python market maker bot on Polymarket Read on Medium

    The author explains how to transition a Python-based market making bot from backtesting to live trading. The guide covers paper trading, live execution, and production monitoring on the Polymarket platform. It is designed for developers interested in prediction market automation.

    Original title: Polymarket Trading Bot : Market Maker: From Backtesting to Live Trading and Production Monitoring

    Read on Medium
  20. MediumPostby Jenny · Zero1 AI

    A non-technical guide to the nine steps of building a trading bot Read on Medium

    The post outlines the nine essential stages required to transform a trading strategy into an active execution loop. The author details the decisions made at each step and the consequences of skipping them. It serves as a conceptual guide for anyone planning to develop trading software.

    Original title: How a Trading Bot Gets Built

    Read on Medium
  21. DEVPostby Odilon HUGONNOT

    Understanding price origins in trading backtests and DeFi oracles through two books Read on DEV

    The author reviews insights from books by Chan and Zhou regarding price data reliability. The post covers critical issues like look-ahead bias in backtesting and the manipulation of flash-loan oracles. It is useful for DeFi developers and quantitative researchers.

    Original title: Backtest and oracle: where does this price come from? Two books read

    Read on DEV
  22. DEVPostby Dron Pancholi

    Why a high win rate over few trades lacks statistical significance Read on DEV

    The author explains why a seventy percent win rate across only twenty trades is statistically meaningless. The post demonstrates how small sample sizes can mimic random chance, making it indistinguishable from a coin flip. This analysis is useful for traders evaluating strategy performance.

    Original title: Why a 70% Win Rate Over 20 Trades Means Nothing

    Read on DEV
  23. DEVPostby Dron Pancholi

    Comparing advertised claims of AI trading bots with their actual performance Read on DEV

    The author investigates the actual performance of grid bots, DCA bots, and AI-based expert advisors. The post contrasts promotional dashboard metrics with findings from independent verifications. This comparison is useful for algorithmic traders and retail investors considering automated tools.

    Original title: AI trading bots in 2026: real returns vs advertised claims

    Read on DEV
  24. DEVPostby Dron Pancholi

    Why every trading model must pass a simple baseline performance test Read on DEV

    The author argues that any trading model must outperform basic benchmarks before being deployed with real capital. The post describes baseline tests such as always-bullish or no-change strategies to evaluate true model value. This guide is helpful for quantitative developers and strategy builders.

    Original title: Beat Nothing First: The Baseline Test Every Trading Model Must Pass

    Read on DEV
  25. DEVPostby Dron Pancholi

    SEBI data shows high losses among retail derivatives traders in India Read on DEV

    The author analyzes a study by SEBI showing that nearly eighty-eight percent of individual futures and options traders lost money in the 2026 fiscal year. The post explores the scale of these financial losses and discusses the psychological factors that drive retail traders to continue. It is relevant for market researchers and retail investors.

    Original title: SEBI’s FY26 Data: 88% of Retail F&O Traders Lost ₹91,685 Crore, And Why They Keep Coming Back

    Read on DEV
  26. arXivPaperby Yupeng Cao, Xueqing Peng, Lingfei Qian et al.

    Introducing FINCOSMOS, an open-world evaluation framework for financial AI agents. Read on arXiv

    The authors present FINCOSMOS, a benchmark designed to test financial LLM agents in realistic, open-world environments. Agents must gather their own web data, construct backtests under specific constraints, and deliver verifiable results without pre-defined datasets. This framework is useful for researchers and developers building autonomous financial assistants.

    Original title: FINCOSMOS: An Open-World Evaluation Paradigm for Agents in Finance

    Read on arXiv
  27. arXivPaperby Zeyu Liu, Xuanzhi Feng, Haoran Xu et al.

    A forecast-augmented policy optimization framework for managing investment portfolios. Read on arXiv

    The authors propose Forecast-Augmented Relative Policy Optimization (FRPO) to improve account-level trading decisions. The system separates asset opportunity forecasting from account-level execution using a jointly learned representation. This research is valuable for quantitative researchers and developers of automated portfolio management systems.

    Original title: Foresee to Trade: Forecast-Augmented Policy Optimization for Portfolio Management

    Read on arXiv
  28. MediumPostby BenjaminCup

    Building a Python backtesting framework for a Polymarket market maker bot Read on Medium

    The author provides a guide on constructing a Python framework to backtest market-making strategies on Polymarket. The post explains how to simulate realistic order execution and optimize bot performance. It is designed for developers interested in prediction markets and algorithmic market making.

    Original title: Polymarket Trading Bot : Market Maker Optimization: How to Backtest Strategies and Tune Trading…

    Read on Medium
  29. MediumPostby Nayab Bhutta

    Analyzing strategy robustness by removing the top five most profitable trades Read on Medium

    The author explores how removing the five most successful trades impacts the overall performance of a backtested strategy. The post discusses how a strategy claiming forty percent annual returns might rely entirely on a few outliers. This analysis helps quantitative researchers evaluate the robustness of their trading systems.

    Original title: What Happens When You Remove the 5 Most Profitable Trades From a Backtest?

    Read on Medium
  30. MediumPostby EA Rebate | Forex Costs & EA Research

    Identifying overfitting and data bias in Expert Advisor backtesting Read on Medium

    The author explains why trading strategies with excellent backtesting results often fail in live trading. The post focuses on detecting overfitting and data bias before deploying an Expert Advisor. It is useful for algorithmic traders looking to improve their strategy validation.

    Original title: Is Your EA Backtest Lying to You?

    Read on Medium
  31. arXivPaperby Henry Zhang

    An open-source bounded LLM research loop for quantitative trading strategy discovery. Read on arXiv

    The authors introduce Autonomous Quant Researcher, a framework that limits LLM actions to prevent overfitting and false discoveries in trading. The model proposes strategies via declarative JSON, while execution and evaluation are handled by trusted, sandboxed code with strict data splits. This tool is highly useful for quantitative researchers looking to automate strategy generation safely.

    Original title: Hard to Fool: A Bounded LLM Research Loop for Quantitative Finance, and What Two Dozen Campaigns Found

    Read on arXiv
  32. MediumPostby Alexzap

    Building custom Python heatmaps for alpha discovery and market risk analysis Read on Medium

    The author provides a guide on creating custom heatmaps using Python. These visualizations are designed to help with alpha discovery, market risk analysis, and evaluating trading strategies. It is aimed at quantitative researchers and developers looking to improve their data visualization tools.

    Original title: Advanced Python Visualizations Every Quant Should Know, Part 6: DIY Heatmaps for Alpha, Risk, and…

    Read on Medium
  33. MediumPostby Daniel Parmeggiani

    Running a production trading system with AI as the sole software engineer Read on Medium

    The author shares their experience of using AI as the only engineer to run a live trading system. The project encountered 49 different points of failure, which could not be resolved by improving the prompts. This post is highly relevant for developers exploring the limits of AI in production environments.

    Original title: I ran a production trading system with an AI as the only engineer

    Read on Medium
  34. DEVPostby Manish Kumbham

    Testing and open-sourcing hundreds of failed crypto trading strategies Read on DEV

    The author tested 748 crypto trading hypotheses and shared the resulting code and analysis. The post details the systematic evaluation of these strategies and documents why they failed. This resource is valuable for crypto developers looking to avoid common algorithmic trading pitfalls.

    Original title: I tested 748 crypto trading strategies and open-sourced the code and autopsy reports of the corpses

    Read on DEV
  35. arXivPaperby Zhan-He Wang, Zi-Yang Lang, Yi-Yang Li et al.

    CosmosAlpha: A graph-based LLM framework for mining predictive alpha factors. Read on arXiv

    The authors introduce CosmosAlpha, a framework that uses multi-relational graph search and Pareto-guided exploration to discover formulaic alpha factors. By analyzing statistical relationships across search branches, the system avoids redundant exploration and balances multiple objectives. This research is useful for quantitative researchers developing alpha generation pipelines.

    Original title: CosmosAlpha: Multi-Relational Graph Search with Pareto-Guided Exploration for LLM-Driven Alpha Mining

    Read on arXiv
  36. arXivPaperby Russlan Ramdowar

    Evaluating the economic trade-offs of running fresh AI stock analysis reports. Read on arXiv

    The authors analyze the cost-benefit ratio of generating new multi-agent AI stock reports versus reusing older analyses with price updates. Using thousands of reports, they test different portfolio configurations under realistic trading and research costs. This study is valuable for investment firms and developers managing AI-driven research budgets.

    Original title: When Is Fresh AI Stock Analysis Worth It? Allocating a Limited Research Budget Across Price Updates, Deep Forecast Refreshes, and Editorial Reviews of Top Stock Picks

    Read on arXiv
  37. arXivPaperby Hao Zhou, Tiejun Ma, Luis Felipe Costa Sperb

    LENS: A stock prediction framework combining LLM priors and news-driven semantic graphs. Read on arXiv

    The authors propose LENS, a framework that integrates static company relationships extracted from LLMs with dynamic news-driven graphs. The system weights the reliability of relations to improve stock return prediction and ranking. This research is useful for quantitative developers working with graph neural networks and alternative data.

    Original title: LENS: Reliability-Aware Stock Prediction via LLM-Extracted Priors and News-Driven Semantic Graphs

    Read on arXiv
  38. MediumPostby BenjaminCup

    A step by step guide to building a Polymarket market maker bot Read on Medium

    This post is the final part of a series on building a market-making bot for Polymarket using Python. The author covers testing, deployment, and monitoring strategies for the automated system. It is designed for developers interested in algorithmic trading on prediction markets.

    Original title: How to Build and Deploy a Polymarket Market Maker Bot in Python: Testing, Monitoring, and…

    Read on Medium
  39. MediumPostby Fabian Weber

    How unrealistic execution prices can distort backtesting results Read on Medium

    The author explains how simulated safety exits can execute at unrealistic prices, inflating backtest results. They share a comparison between a simulated profit of 264,000 dollars and a real-world profit of 5,950 dollars. This post is useful for developers looking to build more realistic execution models in their trading simulators.

    Original title: My Backtest's Safety Exit Sold at Prices That Never Existed

    Read on Medium
  40. arXivPaperby Amir Dehkordi, Alireza Kargarzadeh, Nariman Khaledian et al.

    Testing LLM stock prediction signals under a strict, leak-proof evaluation protocol. Read on arXiv

    The authors introduce OLIVIA, a rigorous testing protocol designed to eliminate look-ahead bias and memorization in LLM-based stock prediction. They apply this protocol to evaluate several decision models reading news about S&P 500 stocks, showing how predictive signals decay on truly blinded test data. This paper is highly valuable for quantitative researchers designing robust backtests.

    Original title: Thinking Fast, Trading Late: System One Decision Models vs the Market

    Read on arXiv
Detail 04 · Morning sheet

The repo of the week, in your inbox

The morning's news and one repository worth knowing, with a short note on why.

How often?

A short issue in the morning, on every day there is news. Written in Hebrew.

Unsubscribe in one click. Your email is used only for the brief.

Past issues