An open-source autonomous agent that runs deep research on web and local documents using any LLM provider, producing cited reports.
TNT-Likely/PanWatch
Self-hosted AI stock monitor for A-shares, Hong Kong and US stocks, with portfolio tracking, price alerts, TradingAgents deep analysis and scheduled reports.
Open-source alternative to:Futu NiuNiu
Yahoo Finance
TradingView
- Stars
- 2,137
- Forks
- 395
- Open issues
- 73
- Last commit
- yesterday
What is it?
PanWatch is a self-hosted web application for monitoring stocks and listed ETFs across mainland China, Hong Kong and the US. Users manage multiple brokerage portfolios, set price alerts that combine conditions with AND/OR logic, and review ranked opportunities and technical indicators such as MACD, RSI and KDJ.
The AI layer runs scheduled pre-market, intraday and daily-report agents, and can start a TradingAgents deep analysis on a holding. That analysis moves through analyst roles, a bull/bear debate, risk review and a portfolio-manager decision. Models can come from any OpenAI-compatible provider or local models via Ollama.
The app is deployed with Docker, offers a paper-trading simulator with T+1 and T+2 settlement rules, and can be installed as a PWA. Notifications go to Telegram, WeCom, DingTalk, Feishu, Bark or webhooks.
Key features
- 01Monitors A-shares, Hong Kong and US stocks plus listed ETFs in one workspace.
- 02Combines price, change, turnover and volume conditions into alert rules with cooldowns and expiry.
- 03Runs scheduled pre-market, intraday and daily-report agents that skip exchange holidays.
- 04Starts TradingAgents deep analysis on a holding, with the reasoning trail kept for review.
- 05Includes a paper-trading simulator with market-specific settlement rules.
- 06Works with any OpenAI-compatible provider or local Ollama models.
Use cases
- Individual investors tracking several brokerage accounts across mainland China, Hong Kong and the US.
- Traders who want threshold alerts pushed to Telegram, Feishu or DingTalk.
- Developers testing LLM-driven trading research workflows on their own infrastructure.
- Users who prefer keeping market data and model keys on a private server.
Topics
FAQ
- What is PanWatch?
- Self-hosted AI stock monitor for A-shares, Hong Kong and US stocks, with portfolio tracking, price alerts, TradingAgents deep analysis and scheduled reports. PanWatch is a self-hosted web application for monitoring stocks and listed ETFs across mainland China, Hong Kong and the US. Users manage multiple brokerage portfolios, set price alerts that combine conditions with AND/OR logic, and review ranked opportunities and technical indicators such as MACD, RSI and KDJ.
- Is PanWatch free to use? What license does it have?
- PanWatch is open source under the MIT license. Check the license terms before using it in commercial products.
- Is PanWatch still maintained?
- Actively maintained. The last commit to PanWatch was on Oct 18, 2026.
- What can I use PanWatch for?
- Individual investors tracking several brokerage accounts across mainland China, Hong Kong and the US. Traders who want threshold alerts pushed to Telegram, Feishu or DingTalk. Developers testing LLM-driven trading research workflows on their own infrastructure. Users who prefer keeping market data and model keys on a private server.
- Where can I get PanWatch?
- The source code and installation instructions are on GitHub: https://github.com/TNT-Likely/PanWatch
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
An open agent framework where the AI agent gets a full Dockerized Linux desktop, a browser, live document editing, and a bridge to your host machine.
An open-source Chinese book on AI Agent design, built around Agent = LLM + context + tools, with runnable companion experiments per chapter.
Open-source web-based control panel for managing Linux servers, websites, Docker containers, databases and AI agents.
AutoGPT is an open-source platform for building, running, and scheduling AI agents that complete multi-step workflows from plain-English descriptions or a visual builder.
Open-source multi-agent LLM framework that simulates a trading firm's analysts, researchers, trader and risk team to reach trading decisions.