Find your next open-source project

Search by name, purpose, or keyword — or paste a GitHub repository URL

ArvinLovegood/go-stock

Finance

A desktop stock analysis app for A-share, Hong Kong and US markets with local data and LLM-driven research tools.

8,200
1,431
Go
GPL-3.0
1 year old

What it does

go-stock is a desktop stock research workbench built with Wails and NaiveUI. It pulls quotes for A-share, Hong Kong and US markets plus on-exchange ETFs, and layers an LLM agent on top for news sentiment, capital-flow and financial analysis, stock screening, and buy/sell alerts. Data is kept locally, and the AI backend can be a hosted API, a local Ollama/LM Studio model, or an OpenAI-compatible gateway.

The feature set is unusually broad for a single desktop tool: K-line charts with indicators and drawing tools, market news and policy feeds, dragon-tiger list (龙虎榜) and hot-money seat tracking, futures positioning, contract (perpetual) quotes, a signal monitor that scans in the background, and replay/backtest pages for prompt templates and AI recommendations. Alerts can go to Feishu or DingTalk.

Quick start

The project ships prebuilt binaries and does not document a build-from-source path in the README. Download the latest package for your platform from the releases page): go-stock-windows-amd64.exe (Windows) or go-stock-darwin-universal (macOS). Launch it, then open the settings page to configure an AI provider (for example OpenAI, DeepSeek, Ollama, or SiliconFlow) and add stocks to your watchlist. If you are new to the app, start with the zero-basics handbook or the quick start guide.

Constraints

Development is centered on Windows 10; other platforms are listed as untested or partially limited. Linux builds exist per the changelog but are not offered as a download option in the README. Some capabilities are gated to paid sponsorship tiers: background signal monitoring is a VIP2 feature, and VIP2 adds automatic syncing of the last 24 hours of market news at startup. The free tier keeps all features but updates are downloaded from GitHub, so you need working access to GitHub.

LLM analysis quality depends entirely on the configured model and prompt. The author explicitly labels the tool as for entertainment/study and states that AI analysis results are for learning and research only, with investment risk borne by the user. The license is GPL-3.0.

HotRepo's take

On the daily Go-language list recorded 2026-11-17, the project placed 12th with 8,188 stars and 29 stars added that day. That is enough to show steady interest in the Go/AI-tooling space, but one snapshot does not establish a trend. The repository's own changelog is the more useful signal: releases have moved from basic quote display toward agent tooling (150+ data tools, skills, MCP, contract trading) in a matter of months, which suggests the tool is growing faster than its documentation and platform support. Readers who want a guided research assistant with local data storage will find a lot here; readers who need a stable, cross-platform trading terminal should weigh the Windows-first development and the sponsorship-gated features.

Written by AI from the project's README, updated . The project's own docs are the source of truth.

You might also like

  1. ZhuLinsen/daily_stock_analysisAn LLM-driven stock analysis system that scores watchlists across six markets and pushes daily decision dashboards.Python71,789
  2. ErTasselli/OpenTerminalRun a local, keyboard-driven market terminal for US stocks, crypto, macro data and a small portfolio tracker.TypeScript2,137
  3. simonlin1212/Vibe-ResearchA local research workbench for China A-share, US and HK stocks, combining market data, AI-assisted company research, backtesting and a personal research library.TypeScript3,738
  4. spovalova/openterminalAn AI-driven market terminal that turns plain questions into live chart, table and option-chain workspaces tied to your own data keys.TypeScript235
  5. Open-Dev-Society/OpenStockA self-hostable stock market dashboard for tracking quotes, watchlists, alerts and company data across eight markets.TypeScript21,317