An open agent framework where the AI agent gets its own Dockerized Linux desktop, browser, and document tools to carry out real work.
TNT-Likely/PanWatch
Self-hosted AI stock monitor for A-shares, Hong Kong and US stocks, with portfolio tracking, price alerts, TradingAgents-based analysis and automated reports.
- Звёзды
- 2,138
- Форки
- 401
- Открытые issues
- 77
- Последний коммит
- 2 часа назад
Что это?
PanWatch is a self-hosted web application for monitoring stocks and listed ETFs across mainland China, Hong Kong and the U.S. It lets you manage multiple brokerage accounts and holdings, set price alerts with combined conditions, and run scheduled pre-market, intraday and closing workflows on exchange trading days.
For deeper research, PanWatch integrates TradingAgents: four analyst roles feed a bull/bear debate, a risk review and a portfolio-manager decision, with the reasoning trail available in the app and through notification channels. The AI layer works with OpenAI-compatible providers or local models served by Ollama.
Reports and alerts can be delivered via Telegram, WeCom, DingTalk, Feishu, Bark or webhooks. The app is an installable PWA, and it ships as a Docker image with optional paper trading and OpenTelemetry export.
Ключевые возможности
- 01Track multiple brokerage accounts, holdings and P&L across A-shares, HK and US markets.
- 02Run TradingAgents deep analysis with bull/bear debate and portfolio-manager decision.
- 03Build price alerts combining conditions with AND/OR logic, cooldowns and expiration.
- 04Schedule pre-market, intraday and closing agents that skip exchange holidays.
- 05Compute technical indicators such as MACD, RSI, KDJ and candlestick patterns.
- 06Send notifications through Telegram, WeCom, DingTalk, Feishu, Bark or webhooks.
Сценарии использования
- Individual investors who want a self-hosted watchlist with rule-based price alerts.
- Traders reviewing multi-market holdings and getting daily automated reports.
- Developers experimenting with LLM-driven stock analysis using local Ollama models.
- Users who want paper trading to test signal-based entries and exits before using real capital.
Темы
Частые вопросы
- Что такое PanWatch?
- Self-hosted AI stock monitor for A-shares, Hong Kong and US stocks, with portfolio tracking, price alerts, TradingAgents-based analysis and automated reports. PanWatch is a self-hosted web application for monitoring stocks and listed ETFs across mainland China, Hong Kong and the U.S. It lets you manage multiple brokerage accounts and holdings, set price alerts with combined conditions, and run scheduled pre-market, intraday and closing workflows on exchange trading days.
- PanWatch можно использовать бесплатно? Какая у него лицензия?
- PanWatch распространяется с открытым исходным кодом под лицензией MIT. Перед использованием в коммерческих продуктах проверьте условия лицензии.
- PanWatch всё ещё поддерживается?
- Активно поддерживается. Последний коммит в PanWatch — 19 окт. 2026 г..
- Для чего можно использовать PanWatch?
- Individual investors who want a self-hosted watchlist with rule-based price alerts. Traders reviewing multi-market holdings and getting daily automated reports. Developers experimenting with LLM-driven stock analysis using local Ollama models. Users who want paper trading to test signal-based entries and exits before using real capital.
- Где скачать PanWatch?
- Исходный код и инструкции по установке — на GitHub: https://github.com/TNT-Likely/PanWatch
Описание создано ИИ на основе README проекта. Статистика GitHub обновляется ежедневно.
Похожие проекты
A-share (China stock market) AI research workstation for market analysis, theme tracking, stock research, portfolio inspection and post-market review, local-first desktop app.
Self-hosted wardrobe manager that tags clothing photos with AI and suggests daily outfits based on weather and occasion.
A self-hosted subtitle system where an LLM agent finds, verifies and installs subtitles for your media library, and translates when no source has one.
An open-source autonomous research agent that plans questions, gathers web and local sources, and writes cited reports with any LLM provider.
Open-source memory layer that turns agent task outcomes into editable Markdown skill files shared across agents and LLMs.