416 lines
9.3 KiB
Markdown
416 lines
9.3 KiB
Markdown
# Scientific Calculator MCP 服务器
|
||
|
||
> 一个生产级别的 Model Context Protocol(MCP)服务器,为 AI 模型提供高级数学计算功能。支持符号数学(SymPy)、数值计算(NumPy/SciPy)、数据分析(pandas)和图像处理。
|
||
|
||
[](https://opensource.org/licenses/MIT)
|
||
[](https://www.python.org/downloads/)
|
||
|
||
## 📋 目录
|
||
|
||
- [快速开始](#快速开始)
|
||
- [安装指南](#安装指南)
|
||
- [配置说明](#配置说明)
|
||
- [使用示例](#使用示例)
|
||
- [工具参考](#工具参考)
|
||
- [常见问题](#常见问题)
|
||
|
||
## 🚀 快速开始
|
||
|
||
### 第一步:安装依赖
|
||
|
||
```bash
|
||
pip install sympy numpy scipy pandas
|
||
```
|
||
|
||
### 第二步:配置 Claude Desktop
|
||
|
||
编辑 Claude Desktop 的配置文件 `claude_desktop_config.json`:
|
||
|
||
**Windows 用户:**
|
||
```json
|
||
{
|
||
"mcpServers": {
|
||
"scientific-calculator": {
|
||
"command": "python",
|
||
"args": ["-u", "F:\\AAchengguoofAI\\cuz_caculat\\mcp_server.py"]
|
||
}
|
||
}
|
||
}
|
||
```
|
||
|
||
**macOS/Linux 用户:**
|
||
```json
|
||
{
|
||
"mcpServers": {
|
||
"scientific-calculator": {
|
||
"command": "python3",
|
||
"args": ["-u", "/path/to/mcp_server.py"]
|
||
}
|
||
}
|
||
}
|
||
```
|
||
|
||
### 第三步:重启 Claude Desktop
|
||
|
||
在 Claude 中就能使用计算工具了。
|
||
|
||
---
|
||
|
||
## 📦 安装指南
|
||
|
||
### 方式 1:通过 pip 安装(推荐)
|
||
|
||
```bash
|
||
pip install mcp-scientific-calculator
|
||
```
|
||
|
||
### 方式 2:从源码安装
|
||
|
||
```bash
|
||
git clone https://github.com/thinkitpossible/CalcMCP
|
||
cd CalcMCP
|
||
pip install -r requirements.txt
|
||
```
|
||
|
||
### 方式 3:本地开发安装
|
||
|
||
```bash
|
||
git clone https://github.com/thinkitpossible/CalcMCP
|
||
cd CalcMCP
|
||
pip install -e .
|
||
```
|
||
|
||
---
|
||
|
||
## ⚙️ 配置说明
|
||
|
||
### 基础配置
|
||
|
||
确保在 `claude_desktop_config.json` 中配置了服务器路径:
|
||
|
||
```json
|
||
{
|
||
"mcpServers": {
|
||
"scientific-calculator": {
|
||
"command": "python",
|
||
"args": ["-u", "path/to/mcp_server.py"],
|
||
"env": {}
|
||
}
|
||
}
|
||
}
|
||
```
|
||
|
||
### 环境变量(可选)
|
||
|
||
```bash
|
||
# 设置日志级别(可选)
|
||
export LOG_LEVEL=DEBUG
|
||
|
||
# Windows PowerShell
|
||
$env:LOG_LEVEL = "DEBUG"
|
||
```
|
||
|
||
---
|
||
|
||
## 💡 使用示例
|
||
|
||
### 1. 符号数学运算
|
||
|
||
**简化表达式:**
|
||
```
|
||
用户:简化 (x**2 + 2*x + 1) / (x + 1)
|
||
工具调用:symbolic_tool
|
||
参数:
|
||
operation: "simplify"
|
||
expression: "(x**2 + 2*x + 1) / (x + 1)"
|
||
|
||
返回:x + 1
|
||
```
|
||
|
||
**求导数:**
|
||
```
|
||
用户:求 sin(x) + x^3 对 x 的导数
|
||
工具调用:symbolic_tool
|
||
参数:
|
||
operation: "derivative"
|
||
expression: "sin(x) + x**3"
|
||
variable: "x"
|
||
|
||
返回:cos(x) + 3*x**2
|
||
```
|
||
|
||
**求积分:**
|
||
```
|
||
用户:求 2*x 的不定积分
|
||
工具调用:symbolic_tool
|
||
参数:
|
||
operation: "integral"
|
||
expression: "2*x"
|
||
variable: "x"
|
||
|
||
返回:x**2
|
||
```
|
||
|
||
**解方程:**
|
||
```
|
||
用户:解方程 x^2 - 5*x + 6 = 0
|
||
工具调用:symbolic_tool
|
||
参数:
|
||
operation: "solve_equation"
|
||
equation: "x**2 - 5*x + 6"
|
||
variable: "x"
|
||
|
||
返回:[2, 3]
|
||
```
|
||
|
||
### 2. 数值计算
|
||
|
||
**线性方程组求解:**
|
||
```
|
||
用户:求解线性方程组:2x + y = 5, x + 3y = 7
|
||
工具调用:numpy_tool
|
||
参数:
|
||
operation: "linear_algebra"
|
||
array_data: [[2, 1], [1, 3]]
|
||
array_data_b: [5, 7]
|
||
|
||
返回:[2.0, 1.0]
|
||
```
|
||
|
||
**矩阵运算:**
|
||
```
|
||
用户:计算矩阵 [[1, 2], [3, 4]] 的行列式
|
||
工具调用:numpy_tool
|
||
参数:
|
||
operation: "determinant"
|
||
array_data: [[1, 2], [3, 4]]
|
||
|
||
返回:-2.0
|
||
```
|
||
|
||
### 3. 科学计算
|
||
|
||
**常微分方程求解:**
|
||
```
|
||
用户:求解 dy/dx = 2*x,初始条件 y(0)=0
|
||
工具调用:scipy_tool
|
||
参数:
|
||
operation: "solve_ode"
|
||
expression: "2*x"
|
||
initial_condition: 0
|
||
|
||
返回:数值解向量
|
||
```
|
||
|
||
**函数积分:**
|
||
```
|
||
用户:计算 ∫sin(x)dx 从 0 到 π
|
||
工具调用:scipy_tool
|
||
参数:
|
||
operation: "integrate"
|
||
expression: "sin(x)"
|
||
limits: "0:pi"
|
||
|
||
返回:2.0
|
||
```
|
||
|
||
---
|
||
|
||
## 🛠️ 工具参考
|
||
|
||
### symbolic_tool(符号数学)
|
||
|
||
提供以下操作:
|
||
|
||
| 操作 | 参数 | 返回 | 说明 |
|
||
|------|------|------|------|
|
||
| `simplify` | expression | str | 简化数学表达式 |
|
||
| `expand` | expression | str | 展开表达式 |
|
||
| `factor` | expression | str | 因式分解 |
|
||
| `derivative` | expression, variable, order | str | 求导数(可指定阶数) |
|
||
| `integral` | expression, variable, lower_limit, upper_limit | str | 求积分(定积分或不定积分) |
|
||
| `limit` | expression, variable, point | str | 求极限 |
|
||
| `solve_equation` | equation, variable | str | 解方程 |
|
||
| `matrix_multiply` | matrix1, matrix2 | str | 矩阵乘法 |
|
||
| `matrix_inverse` | matrix | str | 矩阵求逆 |
|
||
|
||
**使用示例:**
|
||
```json
|
||
{
|
||
"tool": "symbolic_tool",
|
||
"input": {
|
||
"operation": "simplify",
|
||
"expression": "x**2 + 2*x + 1"
|
||
}
|
||
}
|
||
```
|
||
|
||
### numpy_tool(数值计算)
|
||
|
||
提供以下操作:
|
||
|
||
| 操作 | 参数 | 返回 | 说明 |
|
||
|------|------|------|------|
|
||
| `linear_algebra` | array_data, array_data_b | list | 求解线性方程组 Ax=b |
|
||
| `determinant` | array_data | float | 计算矩阵行列式 |
|
||
| `eigenvalues` | array_data | str | 计算特征值 |
|
||
| `eigenvectors` | array_data | str | 计算特征向量 |
|
||
| `statistics` | array_data | dict | 统计信息(均值、标准差等) |
|
||
| `fft` | array_data | list | 快速傅里叶变换 |
|
||
| `correlation` | array1, array2 | float | 计算两个数组的相关系数 |
|
||
| `pandas_read` | file_path, file_type | str | 读取数据文件 |
|
||
| `pandas_describe` | file_path | str | 数据摘要统计 |
|
||
|
||
**使用示例:**
|
||
```json
|
||
{
|
||
"tool": "numpy_tool",
|
||
"input": {
|
||
"operation": "linear_algebra",
|
||
"array_data": [[2, 1], [1, 3]],
|
||
"array_data_b": [5, 7]
|
||
}
|
||
}
|
||
```
|
||
|
||
### scipy_tool(科学计算)
|
||
|
||
提供以下操作:
|
||
|
||
| 操作 | 参数 | 返回 | 说明 |
|
||
|------|------|------|------|
|
||
| `solve_ode` | expression, initial_condition, t_span | str | 求解常微分方程 |
|
||
| `integrate` | expression, limits | float | 数值积分 |
|
||
| `optimize` | expression, initial_guess | str | 函数优化 |
|
||
| `interpolate` | x_data, y_data, x_new | str | 插值 |
|
||
| `fft` | array_data | list | 快速傅里叶变换 |
|
||
| `root_find` | expression, initial_guess | float | 求根 |
|
||
| `curve_fit` | x_data, y_data | str | 曲线拟合 |
|
||
| `special_function` | func_type, value | float | 特殊函数计算 |
|
||
|
||
**使用示例:**
|
||
```json
|
||
{
|
||
"tool": "scipy_tool",
|
||
"input": {
|
||
"operation": "integrate",
|
||
"expression": "sin(x)",
|
||
"limits": "0:pi"
|
||
}
|
||
}
|
||
```
|
||
|
||
---
|
||
|
||
## ❓ 常见问题
|
||
|
||
### Q1:如何添加到 Claude Desktop?
|
||
|
||
**A:**
|
||
1. 打开 Claude Desktop 配置文件夹
|
||
- **Windows**: `%APPDATA%\Claude\`
|
||
- **macOS**: `~/Library/Application Support/Claude/`
|
||
- **Linux**: `~/.config/Claude/`
|
||
|
||
2. 编辑 `claude_desktop_config.json`
|
||
|
||
3. 在 `mcpServers` 字段添加配置
|
||
|
||
4. 保存并重启 Claude Desktop
|
||
|
||
### Q2:提示 "Python not found" 怎么办?
|
||
|
||
**A:** 确保 Python 已正确安装并在系统 PATH 中:
|
||
```bash
|
||
python --version
|
||
# 或
|
||
python3 --version
|
||
```
|
||
|
||
如果没有找到,使用完整路径:
|
||
```json
|
||
{
|
||
"command": "C:\\Python312\\python.exe",
|
||
"args": ["-u", "path/to/mcp_server.py"]
|
||
}
|
||
```
|
||
|
||
### Q3:模块导入错误怎样修复?
|
||
|
||
**A:** 重新安装依赖:
|
||
```bash
|
||
pip install --upgrade sympy numpy scipy pandas
|
||
```
|
||
|
||
### Q4:如何运行测试?
|
||
|
||
**A:**
|
||
```bash
|
||
python verify_production_readiness.py
|
||
```
|
||
|
||
### Q5:如何调试问题?
|
||
|
||
**A:** 使用完整路径并添加日志:
|
||
```json
|
||
{
|
||
"command": "python",
|
||
"args": ["-u", "-u", "path/to/mcp_server.py"]
|
||
}
|
||
```
|
||
|
||
---
|
||
|
||
## 📊 项目结构
|
||
|
||
```
|
||
.
|
||
├── calculator.py # 核心计算模块(3个工具函数)
|
||
├── mcp_server.py # MCP 服务器实现
|
||
├── mcp_protocols.py # MCP 协议定义(可选)
|
||
├── requirements.txt # 依赖列表
|
||
├── setup.py # 打包配置
|
||
├── pyproject.toml # 项目元数据
|
||
├── advanced_math_problems.py # 测试用例
|
||
├── advanced_math_problems.json# 测试数据
|
||
├── verify_production_readiness.py # 验证脚本
|
||
├── README.md # 英文文档
|
||
├── README_ZH.md # 中文文档(本文件)
|
||
├── QUICKSTART.md # 快速入门
|
||
├── DEPLOYMENT.md # 部署指南
|
||
└── .vscode/ # VS Code 配置
|
||
```
|
||
|
||
---
|
||
|
||
## 📝 版本信息
|
||
|
||
- **版本**: 1.0.0(初始版本)
|
||
- **发布日期**: 2025 年 12 月 7 日
|
||
- **Python 版本**: 3.8+
|
||
- **MCP 版本**: 兼容最新版本
|
||
- **许可证**: MIT
|
||
|
||
---
|
||
|
||
## 🔗 相关链接
|
||
|
||
- **GitHub**: https://github.com/thinkitpossible/CalcMCP
|
||
- **PyPI**: https://pypi.org/project/mcp-scientific-calculator/
|
||
- **MCP 协议**: https://modelcontextprotocol.io/
|
||
- **SymPy 文档**: https://docs.sympy.org/
|
||
- **NumPy 文档**: https://numpy.org/doc/
|
||
- **SciPy 文档**: https://docs.scipy.org/
|
||
|
||
---
|
||
|
||
## 📧 反馈与支持
|
||
|
||
如有问题或建议,请提出 Issue 或联系维护者。
|
||
|
||
**GitHub Issues**: https://github.com/thinkitpossible/CalcMCP/issues
|
||
|
||
---
|
||
|
||
**祝您使用愉快!** 🎉
|