mcp-quant-engine
A quantitative finance MCP server providing 24 tools for option pricing, portfolio optimization, risk measurement, fixed income analysis, and utility functions, enabling AI clients to perform professional financial calculations.
README
量化引擎 MCP 服务器 (mcp-quant-engine)
基于 Model Context Protocol (MCP) 的量化金融计算服务器,使用 FastMCP 框架,为 AI 客户端(如 Claude)提供专业的金融数学计算工具。
功能概览
本服务器提供 24 个 MCP 工具,覆盖量化金融四大核心领域:
| 模块 | 工具数量 | 功能 |
|---|---|---|
| pricing.py | 6 | 期权定价(BS模型、隐含波动率、蒙特卡洛、Greeks、二叉树) |
| portfolio.py | 5 | 组合优化(均值方差、有效前沿、Black-Litterman、HRP、绩效指标) |
| risk.py | 5 | 风险度量(历史VaR、参数法VaR、MC VaR、CVaR、最大回撤) |
| fixed_income.py | 4 | 固定收益(债券定价、久期、凸性、Nelson-Siegel曲线) |
| utils.py | 4 | 工具函数(收益解析、矩阵解析、格式化、输入验证) |
安装
# 克隆项目
git clone https://github.com/yourusername/mcp-quant-engine.git
cd mcp-quant-engine
# 安装依赖
pip install -r requirements.txt
使用
直接运行
python server.py
配置 MCP 客户端
在 Claude Desktop 配置文件中添加:
{
"mcpServers": {
"quant-engine": {
"command": "python",
"args": ["path/to/mcp-quant-engine/server.py"]
}
}
}
工具列表
期权定价工具 (pricing.py)
| 工具 | 描述 | 参数 |
|---|---|---|
black_scholes_call |
BS看涨期权定价 | S, K, T, r, sigma |
black_scholes_put |
BS看跌期权定价 | S, K, T, r, sigma |
implied_vol |
隐含波动率(牛顿迭代法) | price, S, K, T, r, option_type |
monte_carlo_option |
蒙特卡洛期权定价 | S, K, T, r, sigma, n_sims, option_type |
option_greeks |
全部Greeks计算 | S, K, T, r, sigma, option_type |
binomial_tree |
二叉树定价(美式期权) | S, K, T, r, sigma, steps, option_type |
组合优化工具 (portfolio.py)
| 工具 | 描述 | 参数 |
|---|---|---|
mean_variance_optimize |
均值方差优化 | returns_str, cov_matrix_str, target_return |
efficient_frontier |
有效前沿计算 | returns_str, cov_matrix_str, n_points |
black_litterman |
Black-Litterman模型 | P_str, Q_str, cov_matrix_str, market_weights_str, tau |
hrp_clustering |
层次风险平价 | returns_str |
portfolio_metrics |
组合绩效指标 | weights_str, returns_str, cov_matrix_str, rf |
风险度量工具 (risk.py)
| 工具 | 描述 | 参数 |
|---|---|---|
var_historical |
历史模拟法VaR | returns_str, confidence |
var_parametric |
参数法VaR | mean, std, confidence |
var_monte_carlo |
蒙特卡洛VaR | returns_str, confidence, n_sims |
cvar |
条件VaR (CVaR/ES) | returns_str, confidence |
max_drawdown |
最大回撤 | prices_str |
固定收益工具 (fixed_income.py)
| 工具 | 描述 | 参数 |
|---|---|---|
bond_price |
债券定价 | face, coupon_rate, ytm, maturity, freq |
bond_duration |
久期计算 | face, coupon_rate, ytm, maturity, freq |
bond_convexity |
凸性计算 | face, coupon_rate, ytm, maturity, freq |
nelson_siegel |
NS收益率曲线拟合 | beta0, beta1, beta2, tau, maturities_str |
工具函数 (utils.py)
| 工具 | 描述 | 参数 |
|---|---|---|
parse_returns |
解析收益序列 | input_str |
parse_matrix |
解析矩阵 | input_str |
format_result |
格式化数值 | value, precision |
validate_inputs |
输入验证 | args_str |
输入格式说明
- 收益序列:逗号分隔的数值字符串,如
"0.01,0.02,-0.01" - 矩阵:分号分隔行、逗号分隔列,如
"0.04,0.01;0.01,0.09" - 权重向量:逗号分隔的数值,如
"0.3,0.4,0.3"
技术栈
- MCP SDK:
mcp.server.fastmcp.FastMCP - 数值计算: NumPy, SciPy
- 数据处理: Pandas
- 优化求解: scipy.optimize.minimize (SLSQP)
- 层次聚类: scipy.cluster.hierarchy
- 统计分布: scipy.stats.norm
理论参考
- Black, F. & Scholes, M. (1973). The Pricing of Options and Corporate Liabilities.
- Cox, J., Ross, S. & Rubinstein, M. (1979). Option Pricing: A Simplified Approach.
- Markowitz, H. (1952). Portfolio Selection.
- Black, F. & Litterman, R. (1991). Global Portfolio Optimization.
- Lopez de Prado, M. (2016). Building Diversified Portfolios that Outperform Out-of-Sample.
- Nelson, C. & Siegel, A. (1987). Parsimonious Modeling of Yield Curves.
- Jorion, P. (2007). Value at Risk: The New Benchmark for Managing Financial Risk.
- Rockafellar, R. & Uryasev, S. (2002). Conditional Value-at-Risk.
项目结构
mcp-quant-engine/
├── server.py # MCP Server 入口
├── mcp_quant_engine/
│ ├── __init__.py # FastMCP 实例创建
│ ├── pricing.py # 期权定价工具(6个)
│ ├── portfolio.py # 组合优化工具(5个)
│ ├── risk.py # 风险度量工具(5个)
│ ├── fixed_income.py # 固定收益工具(4个)
│ └── utils.py # 数学工具函数(4个)
├── README.md
├── SKILL.md
└── requirements.txt
许可证
MIT License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.