Summary

This project offers A-share investors a local, AI-powered news dashboard tracking global industry signals across 12 key sectors. It combats information overload by distilling insights from 100+ authoritative English sources into concise, daily Chinese summaries and translations, requiring zero API keys.

How It Works

A local Python pipeline fetches and filters 100+ global sources. A user-provided LLM then generates distilled Chinese "key points," translations, and source links. A lightweight local server hosts a single-file HTML dashboard displaying these insights, ensuring all data processing remains on the user's machine.

Quick Start & Requirements

Clone the repo, edit llm.config.json for LLM configuration (local Claude claude-cli or OpenAI-compatible API), and run python3 server.py. Access http://localhost:8793 in your browser and click the refresh icon. Requires Python 3.7+ and internet access.

Highlighted Details

• Tracks 12 A-share sectors (AI, Semiconductor, EV, etc.) using 108+ curated global sources.
• AI distillation produces 3-5 concise Chinese "key points" per sector, with deduplication and original text links.
• Dual-language presentation automatically translates English titles to Chinese.
• Fully local execution: All data processing occurs on the user's machine, with no external dependencies or data sharing.
• Flexible LLM integration supports zero-cost local Claude subscriptions or any OpenAI-compatible API.
• Includes built-in keyword filtering for irrelevant content.

Maintenance & Community

Maintained by Simon Lin. Encourages PRs for new sources/sectors. No specific community channels or roadmap links provided.

Licensing & Compatibility

MIT License permits broad use, modification, and distribution, including commercial integration. Emphasizes local-only operation and data residency.

Limitations & Caveats

Provides industry trends, explicitly not financial advice. Users must provide and configure their own LLM access. Some international sources may require a proxy.