
Financekit Mcp
io.github.vdalhambra由 GitHub 帳號 vdalhambra 發布v1.2.0更新於 Oct 24, 2026
Stock quotes, technical analysis, crypto prices, and portfolio insights for AI agents
概覽
讓 AI 助理取得股票與加密貨幣報價、技術指標、市場總覽和投資組合分析。
- 可以做什麼
- FinanceKit 提供十二個金融市場資料工具。可回傳股票報價、公司資訊,以及最多十個代碼的批次報價,並提供加密貨幣價格、熱門幣種、幣種搜尋和依市值排序的幣種清單。技術分析工具計算 RSI、MACD、布林通道、ADX、隨機指標、ATR、OBV 等指標,附帶型態辨識與白話訊號摘要;其他工具涵蓋歷史價格、主要指數總覽、資產比較和投資組合配置分析。
- 適用情境
- 適合助理需要當前市場資料或已計算好的技術指標,而非原始行情來源的情境,例如篩選某檔股票、比較多個資產、查看加密貨幣動能,或分析投資組合的產業集中度。適用於研究與分析對話,不用於下單交易。
- 執行需求
- 以本機程序透過 stdio 執行,可從 PyPI 安裝 financekit-mcp,或用 uvx 啟動;自行架設需要 Python 工具鏈。另有第三方服務提供的託管 HTTP 端點。股票與加密貨幣未聲明需要 API 金鑰;託管方案含付費等級。
安裝
在 SourceWeft 中
- 開啟 儀表板中的 Financekit Mcp,將其新增到工作區。
- 為需要使用其工具的對話啟用該服務。
Desktop only,透過 STDIO。 STDIO 服務會啟動本機處理程序,因此需要 SourceWeft 桌面主機。
其他 MCP 客戶端
參照 儲存庫 中的啟動說明。
README
[PyPI version] [License: MIT] [Python 3.11+] [Glama MCP Server]
FinanceKit MCP
Financial Market Intelligence for AI Agents — real-time stock quotes, technical analysis, crypto data, and portfolio insights via the Model Context Protocol (MCP).
FinanceKit is an MCP server that gives Claude Code, Cursor, Windsurf, and any AI agent instant access to financial market data. No API keys required for stocks and crypto.
Try it now — no install needed: Open FinanceKit in the MCPize Playground — runs in your browser, free tier (100 calls/month)
Use Cases
Here are concrete examples of what you can ask your AI agent once FinanceKit is installed:
- "Run full technical analysis on TSLA and tell me if it's a buy" — Get RSI, MACD, Bollinger Bands, ADX, Stochastic, and pattern detection with a plain-English signal summary
- "Compare AAPL vs MSFT vs GOOGL performance over the last 6 months" — Side-by-side returns, volatility, Sharpe ratio, and max drawdown
- "Analyze my portfolio: AAPL:50, NVDA:20, MSFT:30, AMZN:10" — Total value, sector allocation, concentration risk, and diversification score
- "What are the trending cryptocurrencies right now? Any worth watching?" — Top trending coins from CoinGecko with price data and momentum
- "Give me a market overview — how are the major indices doing?" — S&P 500, NASDAQ, Dow, VIX, and market sentiment in one call
- "Is Bitcoin overbought? Show me RSI and Bollinger Band analysis" — Full technical analysis works on crypto too, not just stocks
Why FinanceKit?
Tools (12)
Stocks
Crypto
Technical Analysis
Market Overview
Portfolio & Comparison
Installation
⭐ Recommended: MCPize (hosted, no setup)
The fastest way to get started. No terminal, no config files, no Python setup — works in any MCP client:
👉 Install FinanceKit on MCPize — Free tier available (100 calls/month)
Or add to your MCP config directly:
Why MCPize?
- ✅ Zero setup — works immediately in Claude Desktop, Cursor, Windsurf, Claude Code
- ✅ Always up-to-date — new features deployed automatically
- ✅ Scales with you — upgrade to Pro ($29/mo) for 10,000 calls + priority + all premium tools
- ✅ Reliable uptime — managed cloud infrastructure
- ✅ Analytics — track how your agents use the tools
See pricing below for all tiers including Team, Business, and Enterprise.
💻 Advanced: Self-hosted (developers)
For those who prefer to run the server locally:
Claude Code CLI
Claude Desktop / Cursor / Windsurf (local)
From PyPI
From source
Smithery
Note: Self-hosted = full feature access but you manage updates, uptime, and infrastructure. For most users, MCPize is the better choice.
Pricing
Annual plans: Get 2 months free (pay for 10, use 12).
Bundle: Combine with SiteAudit MCP for $39/mo (Pro Combo — save 19%).
Data Sources
- Stocks: Yahoo Finance (via yfinance) — free, no API key
- Crypto: CoinGecko API (free tier, 10K calls/month)
- Technical Indicators: Calculated locally using the
talibrary (RSI, MACD, Bollinger Bands, ADX, Stochastic, ATR, OBV)
All data is cached to minimize API calls: quotes (60s), historical data (1h), crypto (2min), company info (24h).
Examples
Technical Analysis Output
Portfolio Analysis Output
Compatible AI Agents
FinanceKit works with any AI agent or IDE that supports the Model Context Protocol:
- Claude Code (CLI) —
claude mcp add - Claude Desktop —
claude_desktop_config.json - Cursor —
.cursor/mcp.json - Windsurf — MCP settings
- Copilot — MCP configuration
- Any MCP client — stdio or HTTP transport
Support this project
If FinanceKit is useful to you, please consider supporting ongoing development:
- 💎 Upgrade to Pro on MCPize — Best way to support + get premium features
- ⭐ Star this repo — Helps other developers find it
- 💖 Sponsor on GitHub — One-time or recurring support
- 🐦 Share on Twitter/X — Tag @ElAgenteRayo
License
MIT
來源:README.md,提交 08da618
工具
0版本歷史
1- v1.2.0最新Sep 16, 2026
評分
已滿足適用檢查項的 50%
該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。
- 近 30 天內可連通25 分
請保持端點能夠回應 MCP 握手;SourceWeft 每 30 天探測一次。
- 已宣告認證方式10 分
請在註冊中心條目中宣告憑證請求標頭,或保持端點能夠回應探測,以便判定其認證方式。
- 已列出工具15 分
請讓伺服器在無憑證時回應 tools/list,或發布可在沙箱中無憑證啟動的軟體包。
- 描述必需15 分
- 圖示10 分
- 文件15 分
- 授權條款10 分