daily_stock_analysis
SHA-256An open-source AI stock analysis system for A/H/US markets that generates daily decision dashboards and pushes them to WeChat Work, Feishu, Telegram, Discord, Slack, or email. Deploy via GitHub Actions for free.
Smart Download
Download Download Version
v3.17.1 · 216.1 MB
Zero-cost AI stock analysis that pushes daily buy/sell signals to your phone—no server required.
Core Features
- AI decision reports: scores, buy/sell signals, risk alerts, catalysts
- Multi-market data: A-shares, Hong Kong, US stocks, ETFs with K-lines, technicals, fund flows, news
- Web UI for manual analysis, agent chat, backtesting, portfolio management
- 15+ built-in agent strategies: moving average, chan theory, wave, trend, etc.
- Multi-channel push: WeChat Work, Feishu, Telegram, Discord, Slack, email
What It Can't Do
- •This tool provides analysis only, not investment advice. Trade at your own risk. 2. You need to obtain AI API keys (Anspire, Gemini, etc.); some free tiers are limited. 3. News source configuration significantly impacts quality; configure at least one search API. 4. Non-trading days are skipped by default; can be forced. 5. Notification channels require your own bot/webhook setup.
Use Cases
- Daily automatic analysis of your watchlist, with recommendations pushed to your phone
- Retail investors who want AI-driven insights without writing any code
Detailed Introduction
Daily Stock Analysis (DSA) is an open-source, AI-driven stock analysis system covering A-shares, Hong Kong stocks, and US stocks. It aggregates real-time market data, K-lines, technical indicators, fund flows, news, and fundamentals, then generates comprehensive AI decision dashboards (scores, buy/sell signals, risks, catalysts). It supports multi-round agent conversations with 15+ built-in strategies (moving average, chan theory, wave, trend, etc.) and pushes reports to WeChat Work, Feishu, Telegram, Discord, Slack, and email. Deployable via GitHub Actions at zero cost, or locally with Docker. Compared to traditional quantitative frameworks like Qlib or Backtrader, DSA lowers the barrier by leveraging LLMs for natural language reasoning and easy configuration, requiring no coding skills for daily stock monitoring. It also integrates a web UI for manual analysis, backtesting, and portfolio management.
Tags
Getting Started
Download installer
Click the button above to download the installer for your system
Install the software
Open the downloaded dmg file, then drag the app to Applications
Fork this repository to your GitHub account
Configure Secrets: AI API Key (e.g., Anspire), stock list, notification webhook
Enable GitHub Actions and manually trigger a test run
- Fork this repository to your GitHub account
- Configure Secrets: AI API Key (e.g., Anspire), stock list, notification webhook
- Enable GitHub Actions and manually trigger a test run
Latest Release Notes
### 发布亮点
- feat: 新增 Alert API MVP，支持告警规则 CRUD、启停、一次性测试以及触发/通知结果查询；首版覆盖 `price_cross`、`price_change_percent`、`volume_spike`，并保持 legacy 配置兼容和响应脱敏。
- feat: 通知网关新增 ntfy 与 Gotify 一等渠道；通知链路补齐 Web 测试、路由、Actions、诊断与文档，并新增默认关闭的进程内降噪能力。
- feat: Windows 桌面安装版接入自动更新安装流程，支持后台下载、用户确认后重启安装，以及 `.env`、数据库和日志等运行时文件备份/恢复。
- improve: 大盘复盘新增概念排行、人气股、涨停池等底层数据源，支持指数涨跌颜色语义配置，并将复盘结果写入分析历史。
- improve: Web 系统设置页支持 `.env` 配置备份导入/导出；通知测试、Agent/通知配置区域新增局部错误兜底，降低桌面端黑屏风险。
- improve: 报告新增 `REPORT_SHOW_LLM_MODEL` 开关，可隐藏聚合、简报、单股和模板报告底部的运行时模型信息。
- improve: Docker 启动入口自动修复挂载目录权限，并在 `logs` 不可写时降级到控制台输出，减少普通部署的手动修复步骤。
- fix: 数据源缺凭据或连接失败时更温和降级，Longbridge / Pytdx 加入冷却，资金流缺失时避免输出高置信买入结论。
- fix: 分析链路兼容 OpenAI-compatible `content_blocks` 响应，归一策略价格字段，并修复大盘复盘结果滚动与历史记录丢失问题。
- fix: 桌面端发版脚本关闭 electron-builder 自动发布，由 Release workflow 统一上传和发布产物，避免 tag 构建在打包完成后失败。
## What's Changed
* Add the Alert API MVP with rule CRUD, enable/disable controls, one-off testing, trigger history, notification result queries, and initial `price_cross`, `price_change_percent`, and `volume_spike` support by @massif-01 in [#1314](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1314)
* Expand the notification gateway with first-class ntfy and Gotify channels, optional in-process noise reduction, aggregated-notification failure isolation, diagnostics, routing, and Actions env mapping checks by @massif-01 in [#1271](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1271), [#1275](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1275), [#1260](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1260), [#1269](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1269), and [#1276](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1276)
* Add the Windows desktop auto-update installation flow and fix release packaging by disabling electron-builder auto-publish during workflow builds by @ZhuLinsen in [#1256](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1256)
* Improve market review coverage with sector rankings, popular stocks, limit-up pool data, index color semantics, analysis-history persistence, and result-panel scrolling fixes by @ZhuLinsen in [#1278](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1278), [#1295](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1295), [#1307](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1307), and [#1270](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1270)
* Add `.env` backup import/export on the Web settings page, add local error boundaries for notification and Agent settings, and allow reports to hide runtime model metadata by @ZhuLinsen in [#1253](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1253), [#1293](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1293), and [#1294](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1294)
* Fix Docker startup and deployment reliability by repairing mounted-directory permissions and falling back to console logging when the logs directory is not writable by @ZhuLinsen in [#1262](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1262) and [#1263](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1263)
* Make data-provider and analysis failures degrade more safely, including optional fetcher initialization skips, Longbridge reconnect cooldowns, Pytdx cooldowns, and lower-confidence output when fund-flow data is missing by @zbl-96 and @ZhuLinsen in [#1249](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1249) and [#1289](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1289)
* Improve report and LLM compatibility by supporting OpenAI-compatible `content_blocks`, normalizing strategy price fields, and increasing the default timeout budget for fundamental aggregation by @ZhuLinsen in [#1284](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1284), [#1264](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1264), and [#1288](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1288)
* Update documentation for the alert center, notification channels, desktop packaging, README presentation, and PR title conventions by @massif-01 and @ZhuLinsen in [#1301](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1301), [#1258](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1258), [#1259](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1259), and [#1308](https://github.com/ZhuLinsen/daily_stock_analysis/pull/1308)
**Full Changelog**: https://github.com/ZhuLinsen/daily_stock_analysis/compare/v3.16.0...v3.17.1
SHA-256 checksum verified
Checksum extracted from GitHub official Release page
SHA256 Checksum
ba6a9a6f1bfa024d40305f1c8342bb7f5562634dd20bddeae8d9db7b358c7eaaThis checksum is extracted from the GitHub Release page. Verify file integrity after download.
All SHA-256 checksums on this platform are extracted from the project's official GitHub Release page, without any modification. You can independently verify them on the GitHub Releases page.
Open Source Transparency
View GitHub SourceUninstall Info
Delete the forked GitHub repository, stop the Docker container, or remove the local project directory to fully uninstall.
No Extra Dependencies
Ready to use after download. No additional runtime required.
Similar Projects
LocalAI
LocalAI is the open-source AI engine to run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required. Drop-in API compatibility with OpenAI, Anthropic, and ElevenLabs.
goose
Goose is a general-purpose AI agent that runs on your machine. Use it for coding, research, writing, automation, and more. It offers a native desktop app (macOS, Linux, Windows), a full CLI, and an API. Works with 15+ providers (Anthropic, OpenAI, Google, Ollama, etc.) and connects to 70+ extensions via the Model Context Protocol. Built in Rust, part of the Linux Foundation.
ollama
Ollama lets you download, run, and manage large language models locally. One command, multiple platforms, endless possibilities.