MatrixCalculator.net

io.github.gamestarry由 GitHub 帳號 gamestarry 發布v0.1.0更新於 Oct 24, 2026

Exact matrix calculations with step-by-step results for six core matrix operations.

發布方自己的服務：Calculator Tools MCP

已驗證Streamable HTTP可網頁執行資料與分析

服務詳細資訊

識別碼: io.github.gamestarry/matrixcalculator-mcp
傳輸: Streamable HTTP
執行環境: Web executable
信任: Verified
發布於: Oct 15, 2026

安全提示

MCP 服務會接收對話中的工具呼叫請求，並可能執行外部操作。啟用前請先了解該服務能做什麼，並只授予它所需的權限。

註冊中心已驗證此條目所屬的 GitHub 帳號 gamestarry。這並不表明該帳號就是產品方。

SourceWeft 的檢查只涵蓋條目是否完整、端點是否回應；不審查伺服器的程式碼，也不審查其工具的行為。

在 README 中新增徽章

向讀者展示該 MCP 服務已收錄於 SourceWeft，徽章連結到本頁。

[![Listed on SourceWeft](https://sourceweft.com/badge/mcp/io.github.gamestarry%2Fmatrixcalculator-mcp.svg)](https://sourceweft.com/mcp/io.github.gamestarry%2Fmatrixcalculator-mcp?utm_source=readme&utm_medium=badge)

概覽

AI 產生的概覽

遠端 MCP 伺服器，執行六種精確矩陣運算，並回傳可呈現或展開的分步計算過程。

可以做什麼: 對同維度矩陣做加法與減法，逐元素給出計算過程
依列乘行內積相乘相容矩陣
計算方陣行列式，含消去步驟與奇異性判斷
以高斯-約當消去法求反矩陣，並用單位矩陣驗證結果
轉置矩形或方陣，並給出每個結果元素的來源位置
保留整數、小數與 1/2 這類分數字串的精確值，而非近似小數
適用情境: 適合需要可靠、精確線性代數結果而非近似值的場合，也適合使用者想看到解題過程時：逐元素加總、消去步驟，或經單位矩陣驗證的反矩陣。
執行需求: 需要一個遠端 Streamable HTTP 端點；目前公開的 V1 端點無需 API 金鑰。若要在本機開發此伺服器本身，需要 Node.js、npm 與 Cloudflare Wrangler。

安裝前請注意

矩陣資料會傳送到 matrixcalculator.net 的第三方託管服務。此端點公開且無身分驗證，未說明依使用者區隔的存取控制或速率保障。README 說明此專案與主站及其計費系統分開維護。

安裝

在 SourceWeft 中

開啟儀表板中的 MatrixCalculator.net，將其新增到工作區。
為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.gamestarry%2Fmatrixcalculator-mcp/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "matrixcalculator-net": {
      "type": "http",
      "url": "https://mcp.matrixcalculator.net/mcp"
    }
  }
}

README

MatrixCalculator.net MCP

A remote Model Context Protocol (MCP) server for exact, step-by-step matrix calculations powered by MatrixCalculator.net.

Remote MCP endpoint

Official MCP Registry name

Current version: 0.1.0

What it does

MatrixCalculator.net MCP gives MCP-compatible AI clients access to six core matrix operations. Results preserve exact fractions and include structured calculation steps that an AI client can present as a short answer or expand into a detailed solution.

Tool	Description
`matrix_add`	Add two matrices with identical dimensions, with one step per entry.
`matrix_subtract`	Subtract the second matrix from the first, with one step per entry.
`matrix_multiply`	Multiply compatible matrices using row-by-column steps.
`matrix_determinant`	Calculate a determinant with row swaps, elimination steps, diagonal entries, and singularity information.
`matrix_inverse`	Calculate an inverse with Gauss-Jordan row operations, singularity detection, and verification.
`matrix_transpose`	Transpose a matrix and return the source position for every result entry.

Exact values

Inputs may be integers, decimals, or fraction strings:

Calculations use exact fraction arithmetic where possible. For example:

rather than converting the result to an approximate decimal.

Example requests

An MCP-compatible AI client can use the server for requests such as:

The MCP server returns structured result data and detailed steps. The connected AI client decides how much of that detail to display to the user.

Tool inputs

`matrix_add`

`matrix_subtract`

`matrix_multiply`

`matrix_determinant`

`matrix_inverse`

`matrix_transpose`

Connect

Use the following Streamable HTTP endpoint in any MCP client that supports remote MCP servers:

No API key is required for the current public V1 endpoint.

Health check

Local development

Requirements:

Node.js
npm
Cloudflare Wrangler

Clone and install:

bash

git clone https://github.com/gamestarry/matrixcalculator-mcp.gitcd matrixcalculator-mcpnpm install

git clone https://github.com/gamestarry/matrixcalculator-mcp.gitcd matrixcalculator-mcpnpm install

Run type checking:

Start the local Worker:

The local MCP endpoint is normally:

Deployment

The production server runs on Cloudflare Workers.

Production MCP endpoint:

MCP Registry

This server is published in the official MCP Registry as:

Registry metadata is stored in server.json.

Project structure

text

src/├─ index.ts├─ lib/│  └─ matrix-values.ts└─ tools/   ├─ matrix-add.ts   ├─ matrix-subtract.ts   ├─ matrix-multiply.ts   ├─ matrix-determinant.ts   ├─ matrix-inverse.ts   └─ matrix-transpose.ts

src/├─ index.ts├─ lib/│  └─ matrix-values.ts└─ tools/   ├─ matrix-add.ts   ├─ matrix-subtract.ts   ├─ matrix-multiply.ts   ├─ matrix-determinant.ts   ├─ matrix-inverse.ts   └─ matrix-transpose.ts

About MatrixCalculator.net

MatrixCalculator.net provides matrix calculators, learning tools, and step-by-step matrix operations on the web.

This MCP project is maintained separately from the main website so MCP development and deployment do not affect the production calculator or its billing system.

來源：README.md，提交 c4f641b

工具

matrix_add

Add two matrices of the same dimensions entry by entry. Useful for exact matrix addition or explaining each entry step by step. Accepts integers, decimals, and fraction strings (for example 1/2); returns the exact sum matrix and a calculation for each position.

輸入參數matrixA必填matrixB必填

matrix_subtract

Subtract matrix B from matrix A when both have the same dimensions. Useful for exact matrix subtraction, including negative values and step-by-step entry calculations. Accepts integers, decimals, and fraction strings; returns each exact difference and its calculation.

輸入參數matrixA必填matrixB必填

matrix_multiply

Multiply A by B when A's column count equals B's row count, including non-square matrices. Useful for exact matrix multiplication and step-by-step row-by-column dot products. Returns exact results plus each factor pair, product, and entry calculation; accepts fractions and decimals.

輸入參數matrixA必填matrixB必填

matrix_determinant

Find the exact determinant of a square matrix using fraction-preserving elimination. Use for determinant questions requiring row swaps, elimination steps, pivot multipliers, sign changes, diagonal products, or singularity checks. Returns the determinant and detailed operation steps.

輸入參數matrix必填

matrix_inverse

Find an exact inverse of a square matrix with step-by-step Gauss-Jordan elimination. Use for inverse questions needing exact fractions, augmented matrices, row swaps, normalization, elimination, and verification that A times A-inverse is identity. Returns the inverse and verification, or null with singular=true if not invertible.

輸入參數matrix必填

matrix_transpose

Transpose a rectangular or square matrix by exchanging rows and columns. Useful for requests to show where each result entry originated. Preserves exact integer, decimal, and fraction inputs and returns the transposed matrix with source-to-result position mappings.

輸入參數matrix必填