An MCP server for stock analysis, offering tools for chip distribution, pattern analysis, trend reversal detection, and market scanning.
This is a Model Context Protocol (MCP) server that provides stock analysis tools for Claude Desktop. The server integrates existing stock analysis features, including chip distribution analysis, pattern analysis, trend reversal detection, and market scanning capabilities.
get-stock-analysis
: Analyze specific stockscompany-fundamental
: Query company fundamental datastart-bull-bear-scan
: Scan the market for bullish and bearish signals (asynchronous task)start-strong-signal-scan
: Scan the market for strong momentum stocks (asynchronous task)start-hot-stock-scan
: Scan the market for top trending stocks (asynchronous task)get-task-status
: Query the status and results of asynchronous tasksmarket-performance
: Today's market performance, query stocks with biggest gains/losses and highest volumeClone the repository:
git clone https://github.com/yourusername/claude-stock-mcp.git
cd claude-stock-mcp
Install dependencies:
npm install
Test the server in development mode:
npm run dev
Or use MCP Inspector:
npm run inspect
Build the project:
npm run build
Start the server:
npm start
Create or edit the Claude Desktop configuration file:
# macOS
open -a "TextEdit" ~/Library/Application\ Support/Claude/claude_desktop_config.json
Add the server configuration:
{
"mcpServers": {
"stock-analysis": {
"command": "npx",
"args": [
"@gabriel3615/claude-stock-analysis-mcp@latest"
],
"env": {
"FMP_API_KEY": "KEY_HERE",
"APLPVANTAGE_API_KEY": "KEY_HERE"
}
}
}
}
Note: You must use absolute paths and replace "yourusername" with your actual username.
Save the configuration file and restart Claude Desktop.
Once the server is configured, you can access the stock analysis features in Claude Desktop using the following approaches:
Example questions:
Claude will use the get-stock-analysis
tool to analyze specific stocks.
Example questions:
Claude will use the appropriate asynchronous task tools, such as start-bull-bear-scan
, and return a task ID. You can use this ID to query task status and results.
How to use asynchronous tasks:
get-task-status
to check the task statusThis asynchronous approach can handle complex market analyses that run for an extended period, avoiding request timeout issues.
fastmcp provides two convenient ways to test:
Using command line mode:
npm run dev
Using web interface:
npm run inspect
These commands will start the respective test environments, allowing you to interact directly with the MCP server without requiring Claude Desktop.
claude-stock-mcp/
├── src/
│ ├── index.ts # MCP server main file
│ ├── analysis/ # Stock analysis related code
│ │ ├── IntegratedAnalysis.ts
│ │ ├── IntegratedAnalysisTypes.ts
│ │ ├── chip/ # Chip distribution analysis
│ │ ├── patterns/ # Pattern analysis
│ │ └── trendReversal/ # Trend reversal analysis
│ ├── finance/ # Financial data related
│ │ ├── Conditions.ts
│ │ ├── Evaluator.ts
│ │ ├── FMPQuery.ts # Financial Modeling Prep API query
│ │ ├── MarketQuery.ts # Market data query
│ │ └── __tests__/ # Test files
│ ├── strategy/ # Strategy analysis
│ │ ├── BreakoutDetector.ts
│ │ ├── BullOrBearDetector.ts
│ │ └── StrategyAnalysisAgent.ts
│ ├── types.ts # Type definitions
│ ├── config.ts # Configuration file
│ └── util/ # Utility functions
│ ├── TaskManager.ts # Asynchronous task management
│ ├── Logger.ts # Log handling
│ └── util.ts # Common utility functions
├── dist/ # Compiled output
├── logs/ # Log file directory
├── package.json
└── tsconfig.json
To avoid console output interfering with Claude Desktop, the project uses a custom logging system:
logs/
directoryMIT
An MCP server for programmatic control of smartscreen.tv displays via HTTP and MCP commands, with YouTube integration.
An MCP server for accessing Bazi (Chinese astrology) data, requiring an API key.
A healthcare-focused RAG server using Groq API and Chroma for information retrieval from patient records.
An MCP server for fetching verifiable random numbers from the drand network.
Provides real-time prices for assets like precious metals and cryptocurrencies.
Provides AI assistants with comprehensive access to a Plex Media Server.
Generates true random coin flips using the random.org API.
Enable AI Agents to purchase anything in a secure way using Fewsats
A TypeScript-based MCP server for interacting with the Etsy API, featuring a simple notes system.
A server for looking up the daily menu at the Hyteria (B1) restaurant.