finance-mcp
A remote MCP server for market data that provides stock quotes, price history, option chains, finance news, and market snapshots, deployable as a Cloudflare Worker.
README
finance-mcp
A focused remote Model Context Protocol (MCP) server for market data, designed for Cloudflare Workers and AI clients such as ChatGPT/Codex.
Tools
stock_quote— latest / near-real-time quote metadata for up to 10 tickersprice_history— OHLCV history with intraday and daily intervalsoption_chain— option expirations, calls, puts, IV, volume and open interestfinance_news— company/ticker news searchmarket_snapshot— S&P 500, Nasdaq Composite, Dow Jones and VIX snapshot
Data sources
The first release uses Yahoo Finance public/unofficial endpoints. Exchange and upstream rules determine actual latency. Treat quotes as near-real-time unless your upstream entitlement explicitly guarantees real-time data. Options data may be delayed; verify execution decisions against your broker/order book.
Local development
npm install
npm run dev
Health check:
http://localhost:8787/health
MCP endpoint:
http://localhost:8787/mcp
Deploy to Cloudflare Workers
npm install
npx wrangler login
npm run deploy
After deployment, Wrangler returns a Workers URL. Your remote MCP endpoint is:
https://<worker-name>.<account-subdomain>.workers.dev/mcp
Use /health to verify the Worker is reachable.
Connect an MCP client
Configure the remote MCP URL ending in /mcp. This project uses Streamable HTTP through Cloudflare's stateless MCP handler.
Example prompts
Get AAPL and NVDA latest quotes.Show AAPL 5-day 5-minute price history.Get the nearest AAPL option chain and identify the highest-volume calls and puts.Find the latest Apple finance news.Give me a US market snapshot.
Roadmap
Next modules can add SEC filings, FRED macro data, institutional holdings, insider transactions, Cboe/OPRA-capable option feeds and broker-grade real-time market data providers.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。