#7777th of 102 in Agent platformsFor finance and business, trading, stocks and crypto, finance and accounting

panwatch

Stock and portfolio monitor for A-share, HK and US markets with AI analysis

Stars
2.5k
License
MIT
Last commit
Nov 2026
Last release
Nov 2026
Language
Python
Stars, last 29 weeks+2,255
25 Apr 2026: 276 stars2 May 2026: 281 stars9 May 2026: 296 stars16 May 2026: 307 stars23 May 2026: 327 stars30 May 2026: 335 stars6 Jun 2026: 359 stars13 Jun 2026: 372 stars20 Jun 2026: 533 stars27 Jun 2026: 625 stars4 Jul 2026: 652 stars11 Jul 2026: 671 stars18 Jul 2026: 698 stars25 Jul 2026: 717 stars1 Aug 2026: 734 stars8 Aug 2026: 765 stars15 Aug 2026: 790 stars22 Aug 2026: 807 stars29 Aug 2026: 820 stars5 Sept 2026: 908 stars12 Sept 2026: 923 stars19 Sept 2026: 1,038 stars26 Sept 2026: 1,814 stars3 Oct 2026: 1,973 stars10 Oct 2026: 2,075 stars17 Oct 2026: 2,147 stars24 Oct 2026: 2,228 stars31 Oct 2026: 2,343 stars7 Nov 2026: 2,429 stars14 Nov 2026: 2,531 stars

276 on 25 Apr 20262,531 on 14 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

PanWatch tracks multi-account portfolios, quotes and price alerts across mainland China, Hong Kong and US markets. Scheduled agents run pre-market, intraday and closing analysis, and a TradingAgents integration runs a multi-agent debate and risk review on a chosen holding. It uses any OpenAI-compatible provider or Ollama, and sends reports to Telegram, WeCom, DingTalk, Feishu, Bark or webhooks.

Who it is for: Self-hosters who trade A-share, HK or US stocks and want AI-assisted monitoring

Strengths

  • Covers A-share, HK and US markets with exchange calendars and holiday-aware scheduling
  • Works with OpenAI-compatible providers or local models through Ollama
  • Six notification channels including webhooks; installable PWA for mobile
  • Optional OpenTelemetry export of agent runs and LLM calls, off by default

Weaknesses

  • ETF holdings, NAV, fees and tracking error are not yet supplied as structured data
  • Paper trading does not model ETF T+0 exceptions, margin, FX or broker fees
  • Bundled exchange calendars need yearly updates; unpublished dates block automatic runs
  • TradingAgents runs take 3-5 minutes and use model quota; cost depends on the model

What it needs

  • no GPU
  • Docker
  • Needs OpenAI-compatible LLM provider or Ollama, Playwright Chromium (downloaded on first start)
  • Models: OpenAI-compatible APIs, Ollama
  • port 8000

What changed

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Also in Agent platforms

See all 102
Also in Agent platforms
RankProjectScore
1AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder Live demo ↗ (opens in a new tab)188.1k starscustom license95 out of 100
2daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps73.1k starsMIT95 out of 100
3LangflowVisual builder for AI agents and workflows, served as API or MCP server156.7k starsMIT94 out of 100
#4deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills85.5k starsMIT93 out of 100
#5DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs Live demo ↗ (opens in a new tab)161.4k starscustom license92 out of 100