広告掲載
MCP.so
ログイン

Scast MCP Server

@davidkingzyb

リポジトリホームページ

基本情報

カテゴリ

その他

ライセンス

MIT

トランスポート

stdio

作者

davidkingzyb

投稿者

DKZ

Scast MCP Server について

Static Code Analysis and Visualization. Convert Code to UML and Flow Diagram and explain by AI.

#scast#static-code-analysis#visualization

Scast MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "scast": {
      "command": "node",
      "args": [
        "/YOUR_INSTALL_DIR/SCAST/mcp/index.js",
        "/YOUR_WORKSPACE/",
        "C:\\Users\\DKZ\\OTHER_ALLOWED_DIR\\"
      ]
    }
  }
}

Scast MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Scast?

Scast is a Model Context Protocol server that provides static code analysis and visualization. It converts source code into UML diagrams, flowcharts, and AST trees, supporting multiple programming languages via TreeSitter. It is intended for developers who want to analyze, summarize, and visualize code structures within an AI client.

How to use Scast?

Clone the repository, run npm install, then configure the MCP server in your AI client with the command node and the path to /YOUR_INSTALL_DIR/SCAST/mcp/index.js followed by one or more allowed workspace directories. The server exposes two tools: scast_analysis for generating diagrams and keyword explanations, and scast_retriever for searching code definitions.

Key features of Scast

  • Converts code into UML diagrams and flowcharts.
  • Supports C#, JavaScript, Python, TypeScript, C, and C++.
  • Uses TreeSitter for parsing and Mermaid/D3 for visualization.
  • Offers two MCP tools: analysis and retrieval.
  • Can be tried online or run locally as an MCP server.
  • Integrates with Ollama for AI-assisted analysis.

Use cases of Scast

  • Analyze a folder of source code and generate AST trees to understand its structure.
  • Retrieve definitions of classes, methods, or fields using keywords from the AST.
  • Produce visual diagrams (UML, flowcharts) to document code architecture.
  • Power an AI assistant with code visualization and retrieval-augmented generation.

よくある質問

1

What languages does Scast support?

It supports C#, JavaScript, Python, TypeScript, C, and C++ by default, with additional languages available via TreeSitter customization.

2

How do I install and configure Scast as an MCP server?

Clone the repo, run `npm install`, then add the configuration block to your MCP client’s settings, specifying the path to `index.js` and allowed workspace directories.

3

Does Scast require an internet connection?

No; the MCP server runs locally. Online demo is available, but the MCP server works offline after installation.

4

What are the transport and authentication requirements?

The README does not specify transport or authentication details; it uses the standard MCP transport via command invocation and file system access to directories you provide.

5

How do I use the `scast_retriever` tool?

Provide the file path of the source code folder, then supply a keyword (class name, method name, or field name) to retrieve the corresponding source code definition.

コメント

「その他」の他のコンテンツ

Blender

ahujasid

Open-source MCP to use Blender with any LLM

23.6K

Yuppsy

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code

Screenpipe

screenpipe

YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context

22.0K

Glasswarp

Glasswarp

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.

Lemon.io

lemon-io

Request dedicated senior developers to work on your project, write job descriptions, and prep technical interviews — all without leaving your Claude chat. Just describe what you need, and Lemon.io MCP does the rest. 1–3

広告掲載