Key takeaways
- An AI model analyses your stock watchlist each day and pushes a report to your chat, instead of you gathering the data by hand.
- It covers A-shares, Hong Kong, US, Japanese, Korean and Taiwanese shares and ETFs, using free data sources by default.
- Ask follow-up questions in a chat page, with 15 built-in strategies, and review past reports in a web workbench.
- Open source under the MIT licence, and it can run on a schedule through GitHub Actions or Docker.
- The catches: it is not investment advice, outputs come from an AI model, and free data sources carry no stability guarantee.
What it is
Daily Stock Analysis describes itself as an AI-model-based analysis system for watchlist stocks in several markets, which analyses them each day and pushes a "decision dashboard" to chat and email services. It is released under the MIT licence, and the code is on GitHub at ZhuLinsen/daily_stock_analysis.
The problem it solves
Following a personal watchlist means collecting prices, technical indicators, news and announcements from separate places, then forming a view. This is a framing of ours rather than the project's wording.
The README lists the data it brings together across markets and ETFs, and has a model summarise it, so a report arrives in your chat each day instead of being assembled by hand.
What you get
- Structured reports: the project lists a core conclusion, score, trend, risk alerts and catalysts for each stock.
- Delivery where you already work: WeCom, Feishu, Telegram, Discord, Slack or email.
- Follow-up questions: multi-turn chat with 15 built-in strategies.
- A web workbench: manual analysis, report history, backtesting and holdings in one place.
- Hands-off running: GitHub Actions, Docker or local scheduled tasks can trigger the analysis.
The catches
- Not financial advice: the README says the project is for learning and research only, warns that stock markets carry risk, and says the author accepts no responsibility for losses.
- AI-generated output: reports come from a language model and include scores and price levels. They can be wrong, so treat them as text to check, not signals.
- Data reliability: free data sources work with no setup, but the README says stability is not guaranteed.
- Setup needs accounts: you must configure at least one AI model service, and the project relies on third-party services for models and news search.
- Maturity: multi-agent orchestration is labelled experimental, and the main README is in Chinese, with English and Traditional Chinese versions linked.
What we did not test
We did not run the project, connect a model or data source, or check any report for accuracy. We did not test the strategies or backtesting, and we did not review how it handles API keys or what data leaves your machine. Every claim here comes from a literal quote in the project's README or the GitHub API, and the feature descriptions are the project's own.
Verdict
Editors' view: this suits developers and hobbyists who already follow a watchlist, want a summary delivered automatically and are comfortable supplying their own model keys. It is a reading aid, not a decision-maker, and nobody should trade on its output without their own checks. The project makes no promise about returns, and neither do we.
Sources
Need help adopting this, or have a software project in mind?
Tell us what you are building. The initial consultation is free. Email [email protected] or follow @zolnaltd.
Start a conversation →