Give Claude, Cursor, Windsurf, or any MCP-compatible AI direct access to live stock data — SEC filings, screeners, insider activity, technicals, and more.
Included with Pro at $19.99/month · Cancel anytime
The Model Context Protocol (MCP) is an open standard that lets AI assistants call external tools and data sources. A stock market MCP server gives your AI the ability to pull live financial data — prices, SEC filings, insider trades, technicals — instead of relying on stale training data.
When you ask your AI to "analyze Apple's financials" or "screen for undervalued stocks," the MCP server fetches current data and returns it directly to the AI's context. The result is investment research grounded in real, up-to-date numbers.
Your AI calls the MCP server behind the scenes to pull live data and answer your questions.
Research a Company
"What's Apple's current valuation? Pull the last 5 years of financials and write me a brief analysis of their financial health."
Screen for Stocks
"Find large-cap stocks with P/E under 20, revenue growth above 15%, and RSI below 40. Then compare the top 5 side by side."
Track Insider Activity
"Show me insider buying activity for NVDA over the last 3 months. Are insiders buying or selling?"
Review Your Portfolio
"Show me my portfolio. For each holding, pull the latest summary and insider activity. Flag any positions where insiders are selling heavily."
Earnings Calendar
"Who reports earnings tomorrow? Do any of them overlap with my watchlist?"
Compare Companies
"Compare MSFT, GOOGL, and AMZN side by side — valuation, growth, and profitability."
Get started in under a minute with a Pro subscription.
Each tool is purpose-built for stock market research. Your AI selects the right one automatically based on your question.
| Tool | Description |
|---|---|
| get_stock_summary | Price, P/E, P/B, EV/EBITDA, market cap, dividend yield, 52-week range |
| get_financial_statements | Multi-year income statement, balance sheet, and cash flow from SEC XBRL filings |
| get_technical_indicators | Moving averages (SMA/EMA), RSI, period returns, volatility, and historical prices |
| screen_stocks | Screen 5,000+ stocks with 30+ filters: valuation, growth, momentum, dividends |
| get_insider_transactions | SEC Form 4 buy/sell data — names, titles, shares, prices, transaction dates |
| get_institutional_holders | Top holders from 13F filings — Berkshire, Vanguard, BlackRock, and 25+ more |
| get_earnings_calendar | Upcoming earnings with report timing and consensus EPS estimates |
| get_market_movers | Top gainers and losers by percentage change |
| compare_stocks | Side-by-side comparison of up to 5 stocks across all metrics |
| get_sector_performance | GICS sector returns and relative performance |
| search_stocks | Search by ticker or company name |
| get_portfolio | Your holdings with cost basis, P&L, and sector allocation |
| get_watchlist | Your watchlist tickers with current prices |
| get_alerts | Your price and earnings alerts |
| get_company_news | Recent headlines and SEC filing links |
Pull real-time stock quotes instead of relying on your AI's training data, which could be months old.
Read real 10-K and 10-Q financial statements instead of hallucinated numbers. Revenue, earnings, and cash flow you can trust.
Use 30+ quantitative filters to find stocks matching your criteria — valuation, growth, momentum, dividends, and more.
See what insiders are buying and selling from Form 4 filings, and which hedge funds hold the stock from 13F filings.
Your AI can access your portfolio, watchlist, and alerts for personalized analysis you can't get anywhere else.
No npm packages, no Docker, no local server. Just a URL and an API key — paste it into your AI client and go.
The difference between AI research with and without live data is the difference between an analyst with a Bloomberg terminal and one working from memory alone.
Three steps to connect your AI assistant to live stock market data.
The MCP server is included with Elite Stock Research Pro at $19.99/month.
View pricing & features →Go to your account page, open the API Keys tab, and click Generate Key. Copy the key — it starts with esr_ and is only shown once.
Open Settings → Developer → Edit Config and add:
{
"mcpServers": {
"stock-research": {
"type": "http",
"url": "https://www.elitestockresearch.com/api/mcp",
"headers": {
"Authorization": "Bearer esr_your_api_key_here"
}
}
}
}Run this in your terminal:
claude mcp add-json stock-research '{"type":"http","url":"https://www.elitestockresearch.com/api/mcp","headers":{"Authorization":"Bearer esr_your_api_key_here"}}'Open Settings → MCP → Add Server, select "URL" type, and enter:
URL: https://www.elitestockresearch.com/api/mcp
Header: Authorization: Bearer esr_your_api_key_here
Open the MCP panel, click Add Server, and enter the same URL and header above.
Add to ~/.codex/config.toml:
[mcp_servers.stock-research]
url = "https://www.elitestockresearch.com/api/mcp"
http_headers = { Authorization = "Bearer esr_your_api_key_here" }A stock market MCP server is a service that exposes live financial data to AI assistants via the Model Context Protocol. It lets your AI pull current stock prices, SEC filings, insider transactions, and more — turning it into a data-driven research assistant instead of one that guesses from training data.
Any MCP-compatible client works, including Claude Desktop, Claude Code, Cursor, Windsurf, and Codex CLI. If your AI tool supports adding MCP servers via URL, it's compatible.
No. This is a remote MCP server — there's nothing to install. You add the server URL (https://www.elitestockresearch.com/api/mcp) and your API key to your AI client's configuration. Setup takes under a minute.
Stock metrics come from major financial data providers. Financial statements are parsed directly from SEC XBRL filings (10-K, 10-Q). Insider transactions come from SEC Form 4 filings. Institutional holdings come from 13F-HR filings. All data is refreshed on a regular schedule to stay current.
Stop asking your AI to guess at stock data. Connect it to Elite Stock Research and get answers grounded in real financial data.
Questions? Reach out at [email protected]