MCP & AI Capabilities

What is MCP?

MCP (Model Context Protocol) is an open protocol proposed by Anthropic for connecting AI models with external data sources and tools. Through MCP, AI assistants can securely access real-time data and call external services without hardcoding API calls.

Stock SDK's companion MCP Server — stock-sdk-mcp — instantly gives your AI assistant professional stock market data capabilities.

Architecture Overview

┌─────────────────────────────────────────────────┐
│         AI Client (Cursor / Claude etc.)         │
│                                                 │
│  "Analyze TSLA's recent MACD trend"             │
└────────────────────┬────────────────────────────┘
                     │ MCP Protocol (JSON-RPC / stdio)
                     ▼
┌─────────────────────────────────────────────────┐
│              stock-sdk-mcp Server               │
│                                                 │
│  69 Tools + 11 Resources + 6 Prompts + 5 Skills │
│                                                 │
│  ┌───────────────────────────────────────────┐  │
│  │           stock-sdk (Core SDK)             │  │
│  │  Quotes / K-line / Indicators / Futures .. │  │
│  └───────────────────────────────────────────┘  │
└─────────────────────────────────────────────────┘

Core Capabilities

69 MCP Tools

Covering all core features of stock-sdk:

• Real-time Quotes: A-shares / HK / US stocks / Funds (smart name search)
• K-line Data: Daily / weekly / monthly K-line, minute K-line, K-line with technical indicators
• Sector Data: Industry / concept sector quotes, constituents, K-line
• Capital: Fund flow, northbound funds, Dragon-Tiger list, block trades, margin trading
• Futures / Options: Domestic & global futures quotes/inventory, ETF / index / commodity options
• Compound Analysis: Full stock analysis, comparison, screening, market overview, sector deep-dive
• Code Lists / Extended: Full-market codes, trading calendar, dividends

👉 View full tool list

7 Resources + 4 Templates

Static resources the AI can read directly (trading calendar, market code lists, sector lists) plus parameterized resource templates (single-stock quote / K-line, sector detail).

👉 View full resource list

5 AI Skills

Skills are scenario-based wrappers around the underlying Tools, using predefined Chain-of-Thought (CoT) to guide AI through professional analysis tasks:

Skill	Description
Stock Technical Analyst	Deep K-line pattern and indicator analysis with buy/sell advice
Smart Stock Screener	Filter 20,000+ stocks across markets by custom criteria
Market Deep Overview	Panoramic scan of indices, sectors, concepts, and sentiment
Portfolio Monitor	Real-time position tracking, anomaly detection, P&L calculation
Smart Money Tracker	Northbound funds + Dragon-Tiger institutions + block trades + fund flow to track main-force capital

👉 View Skills details

6 MCP Prompts

Built-in preset prompts any MCP client can use directly: stock-analyst (technical analysis), stock-screener (screening), market-overview, realtime-monitor (watchlist), smart-money-tracker, futures-overview.

Supported AI Tools

AI Tool	Type	Status
Cursor	IDE	✅ Fully supported
Claude Desktop	Desktop app	✅ Fully supported
OpenClaw	MCP Gateway / AI assistant	✅ Fully supported
Antigravity	VS Code extension	✅ Fully supported
Codex CLI	Terminal tool	✅ Fully supported
Gemini CLI	Terminal tool	✅ Fully supported

👉 View detailed setup guide

Quick Start

Just two steps to give your AI stock data capabilities:

Step 1: Add to your AI tool's MCP configuration:

{
  "mcpServers": {
    "stock-sdk": {
      "command": "npx",
      "args": ["-y", "stock-sdk-mcp"]
    }
  }
}

Step 2: Restart your AI tool and start chatting:

> Analyze Tencent's recent MACD trend. Is there a golden cross?
> Find me the top 10 STAR Market stocks by gain today
> What are the constituents of the AI concept sector?

The AI will automatically call stock-sdk-mcp tools to fetch real-time data and perform analysis.