← 返回 Skills 市场
fredtai

Fintech Research

作者 fredtai · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
39
总下载
0
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install fintech-research
功能描述
100% free, self-hosted equity research toolkit for Claude Code with native Hong Kong stock support. Provides 24 analysis skills across 4 command families (/p...
使用说明 (SKILL.md)

Fintech Research Toolkit

Transform Claude Code into a professional equity research agent with native Hong Kong stock support — all 100% free and self-hosted.

What This Skill Provides

4 Command Families

Command Purpose Trigger Keywords
/probe Thematic discovery, supply-chain scanning, alt-plays "screen", "find ideas", "theme", "discover"
/dive Single-company deep analysis "analyze", "deep dive", "business model", "forensics"
/track Position tracking, thesis validation, event radar "monitor", "watchlist", "track", "thesis check"
/landscape Macro research across CN-HK-US "macro", "yield curve", "trade flows", "rates"

24 Analysis Skills

  • 9 Anthropic official skills (Apache 2.0): initiating-coverage, earnings-preview, earnings-analysis, model-update, morning-note, catalyst-calendar, thesis-tracker, idea-generation, sector-overview
  • 14 Community skills: themes, supply-chain, alt-plays, business-model, financial-forensics, earnings-scorecard, reporting-quality, management, watchlist, thesis-check, event-radar, yield-curve, trade-flows, labor-market

Free Data MCP Server

A local MCP server (mcp-server/server.py) providing 10 tools:

Tool Data Source Free Basis
get_ticker_info yfinance Open source
get_price_history yfinance Open source
batch_ticker_info yfinance Open source
get_ah_spread yfinance Open source
get_hkex_disclosures HKEX News Public access
get_sec_filings SEC EDGAR Official free API
get_macro_data FRED Official free API
get_news_signals NewsAPI + RSS Free tier
fiscal_calendar Local JSON Self-maintained
run_sql SQLite cache Local

HK Equity Support

  • 65 pre-configured .HK tickers covering Hang Seng Top 50 + tech leaders
  • Tencent (0700.HK), Alibaba (9988.HK), Meituan (3690.HK), Xiaomi (1810.HK), BYD (1211.HK), Kuaishou (1024.HK), Li Auto (2015.HK), and more
  • AH spread analysis (A-share vs H-share premium/discount)
  • HKEX regulatory announcements scraping
  • Stock Connect flow context

Setup

# 1. Clone and install dependencies
git clone https://github.com/fredtai/Fintech-research.git
cd Fintech-research
pip install -r requirements.txt

# 2. Start Claude Code
claude

# 3. Verify MCP server is connected
/mcp

# 4. Start researching
/probe AI infrastructure in HK
/dive 0700.HK
/track 0700.HK
/landscape CN-HK-US rates

Optional API Keys

The toolkit works without any API keys for all core features. Optional keys for enhanced data:

Customization

  • .claude/mode.mdnew or experienced
  • .claude/style.md — Control depth (quick/balanced/deep), tone (professional/conversational), coverage (global/hk-only/us-only)

License

Apache 2.0 — Community skills and scaffolding. Anthropic skills bundle is also Apache 2.0 (vendored from anthropics/financial-services).

安全使用建议
Install only if you are comfortable with a local MCP server that can contact public finance/news providers, cache data locally, and let the agent run raw SQL against its cache. Use it in a dedicated workspace, avoid sharing confidential portfolio or unpublished research unless you understand where it will be stored, and treat `run_sql` as a manual-review feature until it is restricted to read-only SELECT queries.
能力标签
cryptorequires-sensitive-credentials
能力评估
Purpose & Capability
The market-data, HK equity research, public API, and optional API-key behavior fits the stated purpose, but `mcp-server/server.py` exposes `run_sql` as caller-supplied SQL while describing it as read-only, and `docs/tools.md` documents arbitrary SQL against the local cache.
Instruction Scope
Many natural-language triggers are broad and can route ordinary finance conversation into tool-using workflows; the initiating-coverage workflow also contains a contradictory path that says to execute missing prerequisite tasks after verification fails, although its example later asks the user first.
Install Mechanism
Installation is ordinary source checkout plus `pip install -r requirements.txt`; dependencies are public packages and I found no hidden post-install script, startup hook, obfuscation, or automatic persistence mechanism.
Credentials
Outbound requests to yfinance, HKEX, SEC, FRED, NewsAPI, and RSS are disclosed and purpose-aligned; `FRED_API_KEY` and `NEWS_API_KEY` are optional and declared, but users should understand ticker/search queries may be sent to third-party services.
Persistence & Privilege
The toolkit creates and updates a local SQLite cache, and some watchlist/thesis instructions mention cross-session storage without a clear storage path or confirmation gate; combined with unrestricted `run_sql`, local research state can be modified or deleted by the agent.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install fintech-research
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /fintech-research 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Fintech-research 2.0.0 is a major upgrade, introducing a comprehensive, self-hosted equity research toolkit for Claude Code with deep support for Hong Kong stocks. - Adds 24 analysis skills across 4 research command families: /probe, /dive, /track, /landscape. - Integrates a local MCP server connecting to free data sources: yfinance, HKEX, FRED, NewsAPI, and RSS. - Provides out-of-the-box support for .HK tickers, AH spread analytics, and HKEX disclosures. - Works fully offline and free, with optional API keys for macro and news data. - Includes advanced customization via mode/style config files for user experience and research depth.
元数据
Slug fintech-research
版本 1.0.0
许可证 MIT-0
累计安装 2
当前安装数 2
历史版本数 1
常见问题

Fintech Research 是什么?

100% free, self-hosted equity research toolkit for Claude Code with native Hong Kong stock support. Provides 24 analysis skills across 4 command families (/p... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 39 次。

如何安装 Fintech Research?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install fintech-research」即可一键安装,无需额外配置。

Fintech Research 是免费的吗?

是的,Fintech Research 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Fintech Research 支持哪些平台?

Fintech Research 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Fintech Research?

由 fredtai(@fredtai)开发并维护,当前版本 v1.0.0。

💬 留言讨论