home / mcp / howrisky mcp server

HowRisky MCP Server

Official Model Context Protocol (MCP) server for AI-powered financial risk analysis. Monte Carlo simulations with fat-tail modeling (CVaR, VaR, ruin probability) for Claude Desktop, Cursor, Windsurf, Cline, Copilot. 8 tools: portfolio risk, startup equity, real estate, Kelly criterion betting. Proprietary KDE algorithm.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "howrisky-howrisky-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "howrisky-mcp-server"
      ],
      "env": {
        "HOWRISKY_API_KEY": "your-api-key-here"
      }
    }
  }
}

HowRisky MCP Server provides Monte Carlo risk analysis capabilities for AI-driven decisions, enabling you to run sophisticated risk simulations across portfolios, real estate, startups, and other assets through a unified MCP interface. You can access metrics like CVaR, VaR, ruin probability, and scenario analyses directly from your AI tools.

How to use

You connect an MCP client to HowRisky to access a suite of risk tools. Start by configuring a local or remote MCP connection, then invoke the available tools from your AI workflow. Your AI can discover available tools, call them with the right parameters, and receive results such as CVaR, ruin risk, survival probability, and percentile timelines. Use the API key you obtain to authenticate requests, and choose the tools that match your risk analysis needs, such as portfolio risk, real estate cash flows, or startup equity modeling.

How to install

Prerequisites: have Node.js and npm installed on your machine.

# Final start command shown in the standard configuration
npx -y howrisky-mcp-server

Configuration and startup details

You unlock HowRisky MCP access by providing an API key and using the standard MCP server configuration. The API key is issued by HowRisky and enables you to perform 100 free calls per month, with higher tiers available.

Standard MCP configuration you will use in your tools looks like this.

{
  "mcpServers": {
    "howrisky": {
      "command": "npx",
      "args": ["-y", "howrisky-mcp-server"],
      "env": {
        "HOWRISKY_API_KEY": "your-api-key-here"
      }
    }
  }
}

Using HowRisky with your AI client

1) Obtain your API key from the HowRisky settings page. 2) Add the standard MCP configuration to your AI tool’s MCP configuration. 3) Restart the AI tool and test a simple query, such as asking to compute risk metrics for a given portfolio. 4) Let your AI discover available tools and call the appropriate ones, such as calculate_portfolio_risk or simulate_future_timelines, then interpret the results shown.

Security and access

Keep your API key secure. Do not share it publicly. Use environment variables to inject the key where needed and restrict access to trusted environments.

Notes on testing

Test by initiating a query like: use HowRisky to calculate the risk of a 60/40 portfolio over 20 years. The AI should identify the HowRisky tools, call the appropriate function, and return metrics such as CVaR, survival probability, and ruin risk.

Available tools

calculate_portfolio_risk

Compute risk metrics such as CVaR, VaR, ruin probability, and survival probability for a given portfolio.

simulate_future_timelines

Project year-by-year portfolio evolution with percentile breakdowns to analyze future risk trajectories.

compare_portfolios

Generate side-by-side risk comparisons across multiple portfolios to highlight relative risk characteristics.

text_to_portfolio

Convert natural language descriptions into concrete asset allocations.

add_startup

Model startup equity scenarios with exit outcomes and risk profiles.

add_real_estate

Analyze real estate cash flows, IRR, and mortgage scenarios.

add_private_asset

Model illiquid assets such as private equity funds and similar investments.

add_gamble

Apply Kelly criterion-based analysis for high-risk betting strategies.