Howrisky
БесплатноНе проверенOfficial Model Context Protocol (MCP) server for AI-powered financial risk analysis. Monte Carlo simulations with fat-tail modeling (CVaR, VaR, ruin probability
Описание
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.
README
Monte Carlo risk analysis for AI agents. Institutional-grade financial modeling with fat-tail distributions and proprietary KDE algorithms.
8 Tools: Portfolio risk (CVaR, ruin probability), startup equity, real estate, Kelly criterion betting, and more.
Compatible with: Claude Desktop, ChatGPT Desktop, Cursor, Windsurf, Cline, GitHub Copilot, VS Code, Codex
Standard Config
{
"mcpServers": {
"howrisky": {
"command": "npx",
"args": ["-y", "howrisky-mcp-server"],
"env": {
"HOWRISKY_API_KEY": "your-api-key-here"
}
}
}
}
Get your free API key at: https://howrisky.ai/app/settings (100 calls/month free)
Getting Started
Step 1: Get your API key from https://howrisky.ai/app/settings
Step 2: Add the standard config above to your AI tool's MCP configuration
That's it! Your AI can now access Monte Carlo risk simulations.
Installation
Claude Desktop
Edit config file:
- MacOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Add the standard config above.
Restart Claude Desktop.
Test it: Ask Claude "Using HowRisky, what's the risk of a 60/40 portfolio?"
ChatGPT Desktop
- Open ChatGPT Desktop Settings
- Go to Apps & Connectors → Advanced settings
- Enable Developer mode
- Add MCP server configuration (use standard config above)
Restart ChatGPT Desktop.
Test it: Ask ChatGPT "Use HowRisky to calculate CVaR for 100% SPY portfolio"
Cursor
Add to Cursor's MCP configuration file:
Use the standard config above.
Cursor supports MCP via VS Code extension compatibility.
Windsurf
Add to Windsurf MCP settings:
Use the standard config above.
Windsurf's MCP integration works similarly to Cursor.
Cline (VS Code)
Via Cline MCP Marketplace:
- Open Cline in VS Code
- Search for "howrisky" in MCP Marketplace
- Click Install
- Enter API key when prompted
Manual Setup:
Add to VS Code Settings → Extensions → Cline → MCP Servers:
Use the standard config above.
GitHub Copilot / VS Code
Add to VS Code settings.json:
Use the standard config above in the MCP servers configuration section.
Remote Server (HTTP)
For custom integrations or web-based AI tools:
Endpoint: https://mcp.howrisky.ai
Authentication: Include X-API-Key header with your API key
Documentation: https://howrisky.ai/mcp/docs
Example:
curl -X POST https://mcp.howrisky.ai \
-H "X-API-Key: your-api-key" \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"tools/list","id":1}'
Available Tools
| Tool | Description |
|---|---|
calculate_portfolio_risk |
CVaR, VaR, ruin probability, survival probability |
simulate_future_timelines |
Year-by-year portfolio evolution with percentiles |
compare_portfolios |
Side-by-side risk comparison of multiple portfolios |
text_to_portfolio |
Natural language → asset allocations |
add_startup |
Startup equity modeling with exit scenarios |
add_real_estate |
Real estate with cash flows, IRR, mortgage analysis |
add_private_asset |
Illiquid asset modeling (PE funds, etc.) |
add_gamble |
Kelly criterion for high-risk betting strategies |
Full documentation: https://howrisky.ai/mcp/docs
Example Usage
Once configured, ask your AI:
"Using HowRisky, calculate the risk of investing $100K in a 60/40 portfolio over 20 years"
The AI will:
- Discover HowRisky tools via
tools/list - Call
calculate_portfolio_riskwith correct parameters - Return CVaR, survival probability, ruin risk, and other metrics
Features
Fat-Tail Modeling - Gaussian models underestimate crash risk by 3-10x. Our proprietary KDE captures reality.
Comprehensive Metrics - 12 risk metrics including CVaR 95/99, VaR, ruin probability, percentiles
Private Assets - Model startups, real estate, PE funds, and high-risk gambles
Tax-Aware - 15+ countries supported (US, GB, DE, FR, IT, ES, JP, AU, CA, etc.)
Custom Scenarios - Override historical data with your own market assumptions
Pricing
| Tier | Calls/Month | Price |
|---|---|---|
| Free | 100 | $0 |
| Developer | 10,000 | $99 |
| Professional | 100,000 | $299 |
| Enterprise | 1,000,000 | $999 |
View pricing: https://howrisky.ai/mcp/pricing
Support
- Issues: https://github.com/howrisky/howrisky-mcp-server/issues
- Docs: https://howrisky.ai/mcp/docs
- Email: [email protected]
License
Proprietary - Copyright © 2025 Diogo Seca / HowRisky.ai
You may use this software to access HowRisky MCP API. Modification and redistribution prohibited. See LICENSE for details.
Установка Howrisky
У этого сервера нет опубликованного пакета — он собирается из исходников. Открой репозиторий и следуй инструкции в README.
▸ github.com/howrisky/howrisky-mcp-serverFAQ
Howrisky MCP бесплатный?
Да, Howrisky MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Howrisky?
Нет, Howrisky работает без API-ключей и переменных окружения.
Howrisky — hosted или self-hosted?
Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.
Как установить Howrisky в Claude Desktop, Claude Code или Cursor?
Открой Howrisky на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Похожие MCP
Fetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
автор: modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
автор: xuzexin-hzCompare Howrisky with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai
