sympy-mcp

概述

Sympy-MCP是一个全面的MCP服务器，通过模型上下文协议为LLM提供强大的符号数学功能访问。它暴露了许多SymPy工具，用于操作数学表达式、解方程、执行微积分运算，甚至处理张量微积分等高级主题。该服务器弥合了LLM和专用计算机代数系统之间的差距，使LLM能够进行准确的符号操作，而不会出现通常影响语言模型数学运算的幻觉问题。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• intro

  Introduces a variable with specified assumptions and stores it

  试试:Introduce a variable x as a real positive number试试:Define variable t as time with real assumption
• intro_many

  Introduces multiple variables with specified assumptions simultaneously

  试试:Introduce variables m, c, and k as real positive constants试试:Define variables x, y, z as complex variables
• introduce_expression

  Parses an expression string using available local variables and stores it

  试试:Parse the expression x^2 + y^2试试:Store the equation m*d^2x/dt^2 + c*dx/dt + k*x = F(t)
• solve_algebraically

  Solves an equation algebraically for a given variable over a given domain

  试试:Solve x^2 - 4 = 0 for x试试:Find the roots of the quadratic equation ax^2 + bx + c = 0
• dsolve_ode

  Solves an ordinary differential equation

  试试:Solve the differential equation dy/dx = x试试:Find the solution to the damped harmonic oscillator equation
• integrate_expression

  Integrates an expression with respect to a variable

  试试:Integrate x^2 with respect to x试试:Calculate the definite integral of sin(x) from 0 to pi
• differentiate_expression

  Differentiates an expression with respect to a variable

  试试:Differentiate x^2 with respect to x试试:Find the derivative of sin(x) * cos(x)
• simplify_expression

  Simplifies a mathematical expression using SymPy's canonicalize function

  试试:Simplify (x^2 - 1)/(x - 1)试试:Reduce the expression (x + y)^2 - (x^2 + 2xy + y^2)
• create_matrix

  Creates a SymPy matrix from the provided data

  试试:Create a 2x2 matrix with elements 1, 2, 3, 4试试:Build a 3x3 identity matrix
• matrix_determinant

  Calculates the determinant of a matrix

  试试:Find the determinant of [[1, 2], [3, 4]]试试:Calculate det(A) for matrix A
• calculate_tensor

  Calculates tensors from a metric (Ricci, Einstein, Weyl tensors)

  试试:Calculate the Ricci tensor for Schwarzschild metric试试:Find the Einstein tensor for the given metric
• create_predefined_metric

  Creates a predefined spacetime metric (e.g. Schwarzschild, Kerr, Minkowski)

  试试:Create the Schwarzschild metric试试:Set up the Minkowski metric for flat spacetime

可对比工具

安装

# 设置项目
git clone https://github.com/sdiehl/sympy-mcp.git
cd sympy-mcp
uv sync

# 安装服务器到Claude Desktop
uv run mcp install server.py

# 运行服务器
uv run mcp run server.py

{
  "mcpServers": {
    "sympy-mcp": {
      "command": "/opt/homebrew/bin/uv",
      "args": [
        "run",
        "--with",
        "einsteinpy",
        "--with",
        "mcp[cli]",
        "--with",
        "pydantic",
        "--with",
        "sympy",
        "mcp",
        "run",
        "/ABSOLUTE_PATH_TO_SYMPY_MCP/server.py"
      ]
    }
  }
}

FAQ

这个MCP服务器支持哪些数学运算？

服务器支持代数方程求解、积分和微分、矢量微积分、广义相对论的张量微积分、常微分方程和偏微分方程、矩阵运算、单位转换等。

这个服务器能处理广义相对论计算吗？

是的，它可以从度规计算张量（里奇张量、爱因斯坦张量、外尔张量），并包含预定义的时空度规，如史瓦西度规、克尔度规和闵可夫斯基度规。要进行广义相对论计算，需要安装einsteinpy库。

sympy-mcp 对比