Agentic-Investor: A Financial Analysis

Agentic-Investor is an MCP server that delivers real-time market data, financial analytics, and sentiment insights to Large Language Models. It combines market movers, ticker analysis, options data, historical data, financial statements, ownership trends, earnings calendars, and technical indicators to empower informed decision making and fast, reliable responses from your AI workflows.

How to use

You will connect your MCP client to Agentic-Investor to request market data, analyses, and indicators. Use the available functions to fetch market movers, get comprehensive ticker reports, filter options data, retrieve historical prices, pull financial statements, review institutional holders and insider trades, check earnings calendars, and gauge market sentiment. You can also enable optional features like technical indicators and intraday data by using the appropriate install options and connection methods.

How to install

Prerequisites: ensure you have Python 3.12 or higher and a runtime environment for your MCP usage. You should also have a MCP client tool available to run MCP connections.

# Install the MCP runtime helper if needed
curl -LsSf https://astral.sh/uv/install.sh | sh

# Quick start: install core Agentic-Investor features
uvx agentic-investor

# Optional: include technical indicators (requires TA-Lib)
uvx "agentic-investor[ta]"

# Optional: include Alpaca intraday data (requires Alpaca API keys)
uvx "agentic-investor[alpaca]"

# All optional features
uvx "agentic-investor[ta,alpaca]"

MCP connection configurations

Agentic-Investor supports both an HTTP MCP connection (remote server) and a local stdio connection (direct runtime). Use the following explicit configurations to connect from your MCP client.

{
  "mcpServers": {
    "Agentic-Investor": {
      "type": "http",
      "name": "agentic_investor_http",
      "url": "http://0.0.0.0:8000/mcp",
      "args": []
    }
  }
}

Local runtime configuration (stdio)

If you prefer running Agentic-Investor locally, use the stdio connection. This starts the server in-process and connects through the MCP interface directly.

{
  "mcpServers": {
    "Agentic-Investor": {
      "type": "stdio",
      "name": "agentic_investor_local",
      "command": "uv",
      "args": ["run", "python", "-m", "agentic_investor.server"]
    }
  }
}

Notes on environment variables and security

You may need API keys and credentials for data sources and models. Commonly required keys include OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, and ALPACA_API_KEY/ALPACA_API_SECRET for Alpaca data. When using debugging features, you can enable verbose logging with DEBUG_LOGGING. Never expose secrets in plain text in logs or shared configs.

Tools and capabilities overview

Agentic-Investor exposes a set of practical tools for data retrieval and analysis. You can query market movers, fetch ticker-specific details, filter options chains, pull price history with adaptive intervals, obtain financial statements, retrieve ownership and earnings history, review insider trades, access earnings calendars, and gauge market sentiment from multiple sources. You can also compute technical indicators such as SMA, EMA, RSI, MACD, and BBANDS when TA-Lib is available.

Security and troubleshooting

Keep API keys and credentials in secure storage and apply least-privilege access for MCP clients. If you encounter connection issues, verify the correct MCP URL or command, ensure the runtime (uvx or Python module) is installed, and check that any required environment variables are set. For debugging, enable detailed logs and review the MCP inspector output to identify bottlenecks or rate-limiting behavior.

Available tools

get_market_movers

Fetch market movers data including top gainers, losers, and most active stocks with support for market sessions and optional filtering.

get_ticker_data

Return a comprehensive ticker report with news, analyst recommendations, and upgrades/downgrades within configurable limits.

get_options

Retrieve filtered options chains with date range, strikes, and option type filters.

get_price_history

Obtain historical OHLCV data with adaptive interval selection to optimize data volume.

get_financial_statements

Fetch income, balance sheet, and cash flow statements with parallel fetching support.

get_institutional_holders

Provide major institutional and mutual fund holders for a ticker.

get_earnings_history

Return historical earnings data with configurable entry limits.

get_insider_trades

Provide recent insider trading activity with configurable trade limits.

get_nasdaq_earnings_calendar

List upcoming earnings announcements with date filtering.

get_cnn_fear_greed_index

Retrieve CNN Fear & Greed Index indicators and historical data.

get_google_trends

Fetch Google Trends relative search interest for specified keywords.

calculate_technical_indicator

Compute technical indicators (SMA, EMA, RSI, MACD, BBANDS) with configurable parameters.